Schedule of FYP Proposal Evaluation (FA25)

1
Committee: 1
Team Members:
Dr. Kashif Ayyub (HEAD)
Dr. Mamuna Fatima
Ms. Yasmeen Khaliq
Mr. Ashfaq Ahmed
Venue: Old Conference room (Faculty Hall)
Remarks: BE ON TIME.

1{"id":971,"project_id":1316,"title":"NextPakistan \u2013 Empowering AI Talent, Solving Real-World Problems","prob":"Next Pakistan addresses the lack of a centralized platform in Pakistan for AI and Data Science opportunities. Currently, students, researchers, and professionals struggle to find reliable information about competitions, hackathons, workshops, or datasets, as these are scattered across different sources. This project solves that problem by providing a single hub where organizations can advertise events, share resources, and connect with talent. It bridges the gap between opportunities and the growing AI community in Pakistan.","description":"Next Pakistan is a web-based platform that connects students, teachers, researchers, and organizations in AI and Data Science. Its main feature is an Advertisement Module, where organizations can post upcoming competitions, hackathons, and events in Pakistan. Students and professionals can view these ads, explore details, and register through provided links. The platform also includes a Dataset Module with local datasets, a Learning & Resources Module for tutorials, and a Community Module that supports discussions and problem-sharing, where users can post real-world AI challenges and get solutions from others. Finally, the Admin Module manages users, advertisements, and content.$$\nIn NextPakistan, the modules are:\r\n(i) User Module(ii) ) Advertisement Module iii) Dataset Module(iv) Learning & Resources Module(v) Community & Networking Module(vi) Admin Module$$\nUser Module \u2013 Manages registration, login, and user profiles.\r\nAdvertisement Module \u2013 organizations post upcoming competitions, hackathons, and events for students and professionals to view and join.\r\nDataset Module \u2013 Provides secure, well-organized local datasets for challenges.$$\nLearning & Resources Module \u2013 Offers tutorials, starter notebooks, and study materials.\r\nCommunity & Networking Module \u2013 Provides forums, discussions, and events to encourage collaboration and inclusivity.\r\nAdmin Module \u2013 Allows administrators to manage users, competitions, datasets, and overall platform settings.$$\n$$\nHackathon Heroes$$\n. Advertisement of Competitions & Events \u2013 Organizations can post and promote AI\/Data Science competitions, hackathons, and workshops happening in Pakistan, making opportunities easier to find.$$\nProblem-Sharing in Community \u2013 Users can share real-world AI or Data Science problems, and others can suggest solutions, fostering collaboration.$$\nIndustry Connection \u2013 Companies and organizations can use the platform to reach skilled students and researchers through posted events or problems, bridging the gap between academia and industry.","comments":"","isDraft":0,"status":2,"created_at":"2025-10-01 21:08:41","updated_at":"2025-10-08 12:38:18","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":976,"project_id":1325,"title":"A Web-Based Digitalization Toolkit for Startup Workflow Automation","prob":"Software startups in Pakistan face major difficulties in managing contracts, invoices, and client communication efficiently. Most rely on manual processes that lead to frequent delays, human errors, and miscommunication with clients. Existing international tools are either too complex, expensive, or not localized to Pakistani startup needs. As a result, startups waste valuable time, struggle to maintain professionalism, and risk losing clients due to poor workflow management. There is a clear need for a localized, easy-to-use, and automated digital toolkit that streamlines contracts, invoicing, scheduling, and communication, while also integrating smart features like reminders and risk analysis. This project aims to solve that real problem by building a tailored web-based solution for Pakistani Software Startups.","description":"Events:\r\nClient Side Events\r\n\t1.\tClient submits requirement (even in Roman English).\r\n\t2.\tAI suggests timeline + milestone payments.\r\n\t3.\tContract draft generated \u2192 finalized with e-signature.\r\n\t4.\tClient receives invoices + meeting reminders.\r\n\t5.\tClient dashboard shows progress, payments, and deployment status.\r\n\r\nAdmin Side Events\r\n\t1.\tAdmin receives client requirement.\r\n\t2.\tAI checks feasibility + assigns risk score.\r\n\t3.\tAdmin finalizes contract and milestones.\r\n\t4.\tSystem manages invoices + schedules meetings.\r\n\t5.\tAdmin dashboard shows income, tasks, risk analysis, and deployment updates.\r\n\r\nExisting AI model with training accordingly.$$\n1. Core Workflow Automation Modules\r\nThe project is built on six core modules that automate the daily operations of startups:\r\nUser Management provides secure login and role-based access.\r\n\r\nInvoice Management creates invoices, sends reminders, and delivers them via email.\r\n\r\nContract Module generates contracts from Roman English input with e-signature support.\r\n\r\nAppointment Scheduler offers calendar-based booking and reminders to avoid meeting clashes.\r\n\r\nEmail Notifications automatically send alerts for invoices, contracts, and schedules.\r\n\r\nDashboard & Analytics gives a real-time summary of income, tasks, clients, and activities.\r\n\r\n2. AI-Enabled Smart Features\r\nTo enhance efficiency, the system integrates four AI-powered smart features:\r\n\r\nContract Generator converts Roman English input into professional contracts.\r\n\r\nRisk Analysis identifies unreliable clients based on delayed payments and past activities.\r\n\r\nSmart Recommendations suggest the next best actions, such as reminders, follow-ups, or rescheduling.\r\n\r\nSmart Scheduler prevents booking conflicts and automatically proposes available time slots.$$\nI shall develop (i) Contract Generation. ( from Roman English input with E signature support) \r\n(ii) Invoice Management(convert roman english into contract)\r\n(iii) Dashboard & Analytics(gives real time summary of income, tasks, client and activities)\r\n(iv) Smart Recommendations\r\n( suggests the next best actions like reminders follow ups and rescheduling)$$\nI shall develop (i) User Management(provides secure login and role based access)\r\n (ii) Risk Analysis(identify unreliable clients based on delayed payments and past activities)\r\n(iii) Smart Schedular(prevent booking conflicts and automatically proposes available time slots)\r\n(iv) Appointment Schedular(offers calendar based bookings and reminders to avoid meeting clashes)\r\n(v) Email Notifications(automatically sends alerts for invoices contracts and schedules)$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-03 17:16:07","updated_at":"2025-10-08 12:42:53","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1005,"project_id":1329,"title":"MediusCare: An AI-Powered Health Monitoring and Assistance System for Diabetes and Hypertension Patients","prob":"Patients suffering from diabetes and hypertension in Pakistan face difficulties in consistently monitoring their vitals, managing medications, and accessing timely medical guidance. Manual record-keeping of blood sugar and blood pressure is error-prone and often neglected due to busy routines, low awareness, or lack of access to doctors. This leads to delayed treatment, poor medication adherence, and higher health risks. Our FYP solves this by providing a centralized AI-powered system that enables patients to log vitals, receive automatic alerts, get dosage recommendations, and connect with doctors and caregivers for timely interventions.","description":"MediusCare is an AI-powered web-based health monitoring platform designed for diabetic and hypertensive patients in Pakistan. Patients can manually log their vitals \u2014 such as blood sugar, blood pressure and food details \u2014 into the system. The platform\u2019s AI modules analyze this data using machine learning models, including LSTM for blood glucose forecasting and SVM\/Logistic Regression for hypertension risk prediction.\r\n\r\nBased on these analyses, the system generates actionable insights, sends automated alerts, and provides personalized medication or insulin dosage recommendations (subject to doctor approval). Doctors can review patient histories, evaluate AI suggestions, and issue digital prescriptions through their dedicated panel, while caregivers have real-time access to emergency alerts and monitoring dashboards.\r\n\r\nAdditional key features include virtual consultations (video and text-based), AI-driven data visualization dashboards, automated reminders, lab report uploads using OCR, and periodic report generation. MediusCare ensures accessibility, affordability, and timely medical intervention \u2014 reducing dependency on manual record-keeping, delayed doctor visits, and inconsistent health monitoring.$$\n1. User Management Module: Handles user registration, login, authentication, and role-based access control for patients, doctors, caregivers, and admins. Includes Firebase authentication and secure session management.\r\n2. Health Data Entry Module: Allows patients to log vitals such as blood sugar, blood pressure, and other health indicators. Supports reminders and manual data input for regular tracking. 3. AI Prediction & Analytics Module: Uses AI\/ML models \u2014 LSTM for glucose forecasting and SVM\/Logistic Regression for hypertension risk prediction. Detects abnormal trends and generates insights.\r\n4. Insulin Dosage Adjustment Module: Provides AI-generated insulin dosage recommendations based on vitals and trends, subject to doctor review and approval. \r\n5. Doctor Panel: Allows doctors to view patient data, review AI suggestions, approve or modify dosage recommendations, and conduct consultations.\r\n6. Caregiver Panel: Enables caregivers to monitor patients\u2019 real-time vitals, receive emergency alerts, and track patient health status (read-only access). \r\n7. Virtual Consultation Module: Supports both **video and text-based** consultations between doctors and patients using integrated Zoom APIs and chat features. Includes appointment scheduling and notes. \r\n8. Alert & Notification System: Sends emergency alerts, medication reminders, appointment reminders, and real-time health notifications to patients, caregivers, and doctors. \r\n9. Data Visualization & Dashboard Module: Displays interactive graphs, charts, and risk indicators for patients and doctors, enabling easy interpretation of trends and analytics. \r\n10. Report & History Module: Generates weekly\/monthly health reports, tracks AI predictions, and maintains the complete history of vitals, consultations, and prescriptions. \r\n11. Admin Panel: Centralized control for managing all users, monitoring system activity, tracking AI logs, handling complaints, and ensuring data security and integrity. \r\n12. Lab Report Management Module (with OCR)** | Allows patients to upload lab reports (PDF\/images). Uses OCR technology to automatically extract and store relevant health data into the system for AI analysis.$$\n1. User Management Module \r\n2. AI Prediction & Analytics Module \r\n3. Caregiver Panel \r\n4. Virtual Consultation Module\r\n5. Report & History Module \r\n6. Admin Panel$$\n1. Health Data Entry Module \r\n2. Insulin Dosage Adjustment Module\r\n3. Alert & Notification System \r\n4. Data Visualization & Dashboard Module \r\n5. Doctor Panel\r\n6. Lab Report Management Module$$\n$$\n$$\n$$\n$$\n","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-08 14:42:24","updated_at":"2025-10-20 12:32:07","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":979,"project_id":1317,"title":"CiviLex: Judiciary Management System with AI-Powered Drafting and Land Revenue Integration","prob":"The system addresses several inefficiencies prevalent in Pakistan\u2019s judicial ecosystem, such as manual and unstructured handling of case records, lack of real-time updates for stakeholders, and the requirement of physical visits for land record verification. It also overcomes delays in the drafting of legal documents and the ineffective dissemination of notifications.\r\n\r\n By integrating AI-powered legal drafting, digitized workflows, and online access to land revenue records, the proposed system significantly reduces the need for in-person visits, accelerates legal documentation processes, and enhances communication and transparency among judges, lawyers, clients, and administrators.","description":"The Judiciary Management System with AI-Powered Drafting and Land Revenue Integration is a comprehensive web-based platform designed to automate and modernize judicial operations in Punjab, Pakistan. The system provides role-based access for admins, judges, lawyers, and clients, ensuring secure and efficient handling of judicial data.\r\n\r\nAdmins manage users, cases, notices, and land revenue records; judges view and update assigned cases, hearings, and decisions; lawyers utilize Google\u2019s Gemini LLM-powered drafting assistant for generating petitions, contracts, and notices with structured textual data; and clients can view case details, verify property ownership, and receive updates online.\r\n\r\nThe system replaces manual workflows and bulky PDF-based records with digitized, structured textual data, enabling real-time updates, AI-assisted drafting, and centralized communication among stakeholders. It also integrates land revenue databases and auction management, while providing real-time notifications and analytics dashboards for improved transparency, speed, and accessibility across the judiciary system.$$\n1. Admin Management Module\r\n\r\nManage judges, lawyers, and clients.\r\n\r\nAssign and monitor cases.\r\n\r\nUpload notices, and land revenue records.\r\n\r\nGenerate real-time analytical reports on case progress.\r\n\r\nMaintain structured textual data instead of PDF-heavy storage.\r\n\r\n2. Judge Panel Module\r\n\r\nView assigned cases with full case details.\r\n\r\nUpdate hearing outcomes and case status.\r\n\r\nAccess previous case histories and uploaded documents.\r\n\r\nReceive real-time notifications for hearings and committee updates.\r\n\r\nGenerate case summaries and textual reports.\r\n\r\n3. Lawyer Panel Module\r\n\r\nManage and track assigned cases.\r\n\r\nUse Gemini LLM-based AI drafting to generate legal petitions, contracts, and notices.\r\n\r\nUpload\/download case documents.\r\n\r\nAccess land revenue data and schedules.\r\n\r\nReceive system alerts for hearings, updates, and committee announcements.\r\n\r\n4. Client Portal Module\r\n\r\nRegister and initiate cases through detailed textual intake forms (CNIC, contact, case summary).\r\n\r\nView case progress, judgments, and hearing schedules.\r\n\r\nAccess land revenue and property verification online.\r\n\r\nGet SMS, email, and in-app notifications for updates.\r\n\r\nDownload documents and participate in online auctions.\r\n\r\n5. AI Legal Drafting Module\r\n\r\nIntegrates Google\u2019s Gemini LLM (2.5 Pro) for smart legal drafting.\r\n\r\nGenerates petitions, contracts, and notices in structured text format.\r\n\r\nAllows editing, exporting, and textual formatting for readability.\r\n\r\nSupports legal reasoning, document summarization, and drafting optimization.\r\n\r\n6. Land Revenue Integration Module\r\n\r\nLinks to provincial land record databases for property verification.\r\n\r\nAllows clients and lawyers to view, verify, and download land revenue records.\r\n\r\nDisplays ownership, location, and revenue details through authenticated access.\r\n\r\nReduces manual visits and delays in verification.\r\n\r\n7. Notification & Communication Module\r\n\r\nCentralized real-time notification system for all stakeholders.\r\n\r\nAlerts via SMS, email, and in-app for hearings, and committee updates.\r\n\r\nEnsures immediate communication between judges, lawyers, and clients.\r\n\r\nIntegrates with the dashboard for visual alerts and summaries.$$\nAdmin Management Module, Judge Panel Module, Client Portal Module, and Notification & Communication Module. This includes complete backend development for user management, case tracking, client interface, and real-time notification system implementation.$$\nLawyer Panel Module, AI Legal Drafting Module, and Land Revenue Integration Module. This involves developing the lawyer interface, AI chatbot integration for legal document drafting, and land revenue database connectivity$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-04 22:31:01","updated_at":"2025-10-08 12:40:02","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":963,"project_id":1337,"title":"AI-Powered Mannequin-Based Medical Training System (AIMMS)","prob":"This FYP addresses the critical problem of limited access to affordable and effective medical training equipment in Pakistan. Many medical institutions struggle with the high cost and scarcity of advanced training tools such as laryngoscopes, endoscopy tubes, and CPR mannequins. Traditional mannequins lack interactive features like real-time visualization and feedback, which are essential for developing practical clinical skills. Imported simulators, while advanced, are often prohibitively expensive for local use. This project, the AI-Powered Mannequin-Based Medical Training System (AIMMS), offers a cost-effective, locally developed solution combining hardware and AI technology. It provides real-time anatomy visualization, interactive CPR compression feedback, and detailed performance tracking. By enhancing hands-on training through affordable, interactive tools, AIMMS aims to improve medical education quality, better prepare students for clinical practice, and ultimately contribute to improved healthcare outcomes in resource-limited settings.","description":"The AI-Powered Mannequin-Based Medical Training System (AIMMS) is designed to enhance medical education by providing an affordable, interactive training platform tailored for resource-constrained environments like Pakistan. The system integrates hardware components such as mini cameras, LEDs, and pressure sensors embedded within medical training tools\u2014a laryngoscope, endoscopy tube, and CPR mannequin\u2014to capture real-time data during medical procedures and simulations.\r\n\r\nAt its core, AIMMS captures live video feeds from the embedded cameras inside the laryngoscope and endoscopy tube, which are enhanced and processed using OpenCV for noise reduction and brightness adjustment, ensuring clear visualization in low-light conditions. These feeds are then analyzed using an AI model based on YOLOv8, trained on specialized medical datasets to detect and label critical anatomical structures like the larynx, trachea, and esophagus. This real-time AI-powered feedback helps trainees accurately identify anatomy during practice, bridging the gap between theory and hands-on experience.\r\n\r\nSimultaneously, the CPR mannequin is equipped with load-cell pressure sensors that measure chest compression depth during resuscitation training. Sensor data is collected through an Arduino microcontroller and processed in Python to evaluate the quality of compressions, categorizing them as correct, shallow, or deep. This feedback is displayed interactively via a desktop GUI developed in PyQt, enabling trainees to adjust their technique in real time.\r\n\r\nAll training sessions, including anatomy detection accuracy and CPR performance, are logged into an SQLite database, allowing instructors to monitor trainee progress and generate detailed performance reports.\r\n\r\nThe system integrates hardware, AI, and software modules into a cohesive desktop application that supports live video visualization, anatomy labeling, CPR feedback, and analytics. By delivering a cost-effective, interactive, and locally developed training solution, AIMMS aims to improve medical training accessibility and quality, helping students gain essential clinical skills efficiently and confidently.$$\nModule 1: Hardware Integration\r\nThis module embeds mini cameras and LED lights into medical tools like the laryngoscope and endoscopy tube for real-time visualization. The CPR mannequin is fitted with load-cell pressure sensors to measure chest compression depth. An Arduino microcontroller collects sensor data and communicates with the software, enabling live data capture essential for interactive training.\r\n\r\nModule 2: Video Input & Enhancement\r\nUsing OpenCV, this module captures and synchronizes live video feeds from embedded cameras. It applies noise reduction and brightness adjustment to enhance video quality, especially in low-light conditions, providing clear visuals for trainees during practice.\r\n\r\nModule 3: AI-Based Anatomy Detection\r\nThis module uses YOLOv8 deep learning models for real-time detection and labeling of anatomical structures such as the larynx, trachea, and esophagus. Models are trained on medical datasets like Kvasir and MICCAI and run on PyTorch, offering interactive anatomical feedback to trainees.\r\n\r\nModule 4: CPR Feedback System\r\nLoad-cell sensors in the CPR mannequin measure compression depth. Data is sent via Arduino to a Python app, which classifies compressions as correct (5\u20136 cm), shallow (<5 cm), or deep (>6 cm). Real-time feedback is shown in the GUI, helping trainees improve technique immediately.\r\n\r\nModule 5: Data Processing & Logging\r\nAll training data, including anatomy detection accuracy and CPR metrics, are stored in an SQLite database. This module supports performance tracking by logging user ID, session date, training type, and accuracy, enabling instructors to review trainee progress.\r\n\r\nModule 6: Desktop GUI Application\r\nBuilt with PyQt, this GUI integrates all features into an easy-to-use interface. Tabs include:\r\nLive Video: Shows raw and enhanced camera feeds.\r\nAnatomy Detection: Displays AI-labeled video.\r\nCPR Training: Visualizes compression feedback with depth indicators.\r\nAnalytics: Presents historical performance and progress reports.\r\n\r\nModule 7: Communication & Middleware\r\nHandles serial communication between Arduino and Python using pySerial. It ensures synchronized data exchange and manages errors like hardware disconnections to maintain stable operation.\r\n\r\nModule 8: System Integration & Deployment\r\nThis final module combines all components into one deployable desktop app using PyInstaller. It includes testing for various scenarios, such as low light and different compression speeds, ensuring robustness and usability.$$\nDevelop Module 1: Hardware Integration\r\nEmbed mini cameras, LEDs, and pressure sensors into laryngoscope, endoscopy tube, and CPR mannequin.\r\nWork on Arduino-based data acquisition and hardware communication.\r\n\r\nLead Module 6: Desktop GUI Application development\r\nDesign and implement the user interface using PyQt.\r\nIntegrate live video feeds and CPR feedback visualization.\r\nCreate interactive elements for real-time sensor data display and user interaction.\r\nEnsure synchronization between hardware data and software visualization.\r\nFocus on providing a user-friendly and responsive training platform for medical students.\r\nCollaborate with team members for smooth hardware-software integration.$$\nLead development of Module 3: AI-Based Anatomy Detection\r\nImplement YOLOv8 models for real-time anatomical structure detection and labeling.\r\nTrain and fine-tune AI models using medical datasets like Kvasir and MICCAI.\r\nIntegrate PyTorch framework for efficient model deployment.\r\n\r\nDevelop Module 5: Data Processing & Logging\r\nDesign the SQLite database to store trainee performance data.\r\nManage logging of anatomy detection accuracy and CPR compression metrics.\r\nEnable generation of performance reports and progress tracking for instructors.\r\nSupport optimization of AI models and data handling for real-time feedback.\r\nCollaborate with team members for integration of AI modules with video and hardware components.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-09-30 16:09:30","updated_at":"2025-10-08 12:44:54","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1000,"project_id":1314,"title":"Lodge Logic: Comprehensive Guest House Management System","prob":"Problem:\r\nGuest houses commonly struggle with double bookings, poor room allocation, manual billing, and weak staff coordination because they lack a proper centralized management system. As a result, they face frequent operational delays, financial errors, and unsatisfied guests, ultimately affecting their reputation and revenue.\r\n\r\nSolution (Lodge Logic):\r\nLodge Logic solves these issues by providing a centralized, automated management platform that streamlines the entire guest house workflow \u2014 including reservations, check-in\/check-out, housekeeping updates, billing automation, and detailed reporting. This leads to smooth operations, better financial tracking, and an improved guest experience.","description":"Lodge Logic vs Booking.com \/ Airbnb\r\n-Booking.com & Airbnb lack Urdu\/Roman-Urdu chatbot support \u2192 Lodge Logic offers localized AI chatbot for better accessibility.\r\n-Their focus is on hotels and global users \u2192 Lodge Logic is built specifically for small guest houses and local travelers.\r\n-They rely on international payment methods \u2192 Lodge Logic integrates local payments like JazzCash\/Easypaisa.\r\n-Bookings from multiple channels can cause overlaps \u2192 Lodge Logic provides real-time booking sync to avoid double bookings.\r\n-Their reporting tools are generic \u2192 Lodge Logic gives daily PDF income reports and AI insights for owners.\r\n\r\nLodge Logic \u2013 Project Overview(Working Flow)\r\n-Owner Module\r\nOwner registers and verifies their account.\r\nAdds guest house details, rooms, pricing, and policies.\r\nAssigns admins and monitors real-time room status and reports.\r\n-Admin Module\r\nManages room assignments, check-ins, and check-outs.\r\nHandles housekeeping coordination and customer complaints.\r\nGenerates and manages operational reports.\r\n-Customer Module\r\nCustomers register and search guest houses by location, price, or amenities.\r\nView property details and real-time room availability.\r\nBook rooms instantly and pay via cards, banking, or local payment methods.\r\nReceive instant booking confirmations and notifications.\r\n-Booking & Operations\r\nSystem locks rooms after booking to prevent double bookings.\r\nAdmin updates check-in\/out; housekeeping is notified automatically.\r\nRooms are cleaned and made available again efficiently.\r\n-Billing & Reporting\r\nSystem automatically generates bills and digital receipts at checkout.\r\nOwners access daily, weekly, or monthly income and occupancy reports in PDF format.\r\n-AI & Smart Features\r\nAI Chatbot (GPT) answers guest queries in Urdu\/Roman-Urdu.\r\nSentiment Analysis labels guest reviews as Positive, Neutral, or Negative.\r\nSmart Recommendation System suggests add-ons and upgrades.\r\nFraud Detection (AWS\/Azure) identifies suspicious booking patterns.\r\n In short:\r\nOwner sets up \u2192 Admin manages \u2192 Customer books \u2192 System automates operations & billing \u2192 AI boosts support & insights \u2192 Owner tracks reports.$$\ni. Guest House Owner Modules:\r\n 1. Owner Registration & Login \u2013 \r\nSecure sign-up\/login with identity verification (email, phone,KYC).\r\n 2. Guest House Registration \u2013 Submit property details, photos, amenities, policies.\r\n 3. Room Management \u2013 Add\/remove rooms, manage availability, set prices.\r\n 4. Review & Rating Management \u2013 See guest feedback and reply to reviews.\r\n 5. Reports & Analytics (Basic) \u2013 Booking history, occupancy stats, revenue summaries of different guest houses.\r\n 6. Add an admin for a particular guest house to manage that guest house.\r\n\r\nii. Customer Modules:\r\n 1. User Registration & Login \u2013 Secure sign-up\/login with email, phone, or social accounts.\r\n 2. Guest House Search & Discovery \u2013 Search\/filter by location, price, amenities.\r\n 3. Property Details Page \u2013 Full details including images, policies, and reviews.\r\n 4. Booking & Reservation System \u2013 Select dates, finalize booking, get confirmation.\r\n 5. Payment & Billing \u2013 Secure checkout with multiple payment options and receipts.\r\n 6. Booking History & Tracking \u2013 View upcoming, past, or canceled bookings.\r\n 7. Review & Feedback \u2013 Leave reviews, ratings, and photos after a stay.\r\n 8. Wishlist \/ Favorites \u2013 Save guest houses for future bookings.\r\n\r\niii. Guest House Admin Module:\r\n 1. Admin Dashboard \u2013 Centralized overview of system stats and activities.\r\n 2. User Management \u2013 Approve or block users, manage KYC verification.\r\n 3. Booking Management \u2013 View, approve, or reject bookings; track booking history.\r\n 4. Payment Management \u2013 View payments, earnings summary, request payouts.\r\n 5. Complaint & Support Handling \u2013 Manage user complaints and assign tickets.\r\n 6. Content Management (CMS) \u2013 Manage FAQs, policies, announcements, promotions.\r\n\r\niv. Shared Modules:\r\n 1. Notification System \u2013 Push\/email alerts for bookings, payments, offers.\r\n 2. Profile Management \u2013 Update personal details and preferences.\r\n 3. Authentication & Security \u2013 Secure login, JWT\/session management, password recovery.\r\n 4. Help & Support \u2013 FAQs, contact options, and manual live chat support.\r\n\r\nAI & Smart Features (with apis to be used)\r\n(i)AI Chatbot Concierge (GPT API)\r\n Uses GPT API with NLP to interact with guests in Urdu\/Roman-Urdu, answering queries about room availability, facilities, cancellations, and FAQs \u2014 providing instant 24\/7 support.\r\n\r\n(ii)Guest Review Sentiment Analysis (GPT API)\r\n Automatically analyzes and classifies guest feedback as Positive, Neutral, or Negative using GPT API, helping owners quickly understand customer satisfaction trends.\r\n\r\n\r\n(iii)Smart Recommendation System (Booking.com \/ Expedia \/ Amadeus APIs)\r\n Leverages travel APIs to suggest personalized add-ons, upgrades, or nearby attractions based on user preferences and behavior, enhancing the overall booking experience.\r\n\r\n(iv)Fraud\/Abnormal Booking Detection (AWS \/ Azure AI Services)\r\nUses AWS Fraud Detector or Azure Anomaly Detector API to automatically analyze booking data and detect suspicious or unusual activity \u2014 such as multiple bookings from the same IP address, fake accounts, or abnormal payment patterns. The module flags these bookings for review, helping prevent fraud and protect system integrity.$$\nI shall develop these modules: \r\n\r\n(i)AI Chatbot Concierge (GPT API)\r\n Uses GPT API with NLP to interact with guests in Urdu\/Roman-Urdu, answering queries about room availability, facilities, cancellations, and FAQs \u2014 providing instant 24\/7 support.\r\n\r\n(ii)Guest Review Sentiment Analysis (GPT API)\r\n Automatically analyzes and classifies guest feedback as Positive, Neutral, or Negative using GPT API, helping owners quickly understand customer satisfaction trends.\r\n(iii)Smart Recommendation System (Booking.com \/ Expedia \/ Amadeus APIs)\r\n Leverages travel APIs to suggest personalized add-ons, upgrades, or nearby attractions based on user preferences and behavior, enhancing the overall booking experience.\r\n\r\n(iv)Fraud\/Abnormal Booking Detection (AWS \/ Azure AI Services)\r\nUses AWS Fraud Detector or Azure Anomaly Detector API to automatically analyze booking data and detect suspicious or unusual activity \u2014 such as multiple bookings from the same IP address, fake accounts. The module flags these bookings for review, helping prevent fraud and protect system .$$\nIn lodgelogic I shall develop these modules:\r\n(i) Owner Modules: \r\nRegistration, guest house management, room & booking management, payments, reviews, and basic reports.\r\n(ii) Customer Modules: \r\nRegistration, search & book guest houses, payments, booking history, reviews, and wishlist.\r\n(iii)Employee\/Admin Modules: Dashboard, user & guest house management, booking oversight, payments, complaints handling, and system reports.\r\n(iv) Shared Modules: \r\nNotifications, profile management, security, and help\/support$$\n$$\n$$\nA generic module to add virtual guest houses$$\nAdmin module to handle specific guest house$$\nAI module to handle user qeureis.","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-08 12:47:49","updated_at":"2025-10-20 12:29:27","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1009,"project_id":1347,"title":"UniLearn AI\u2014university + learning powered by AI.","prob":"The Smart University Learning Platform enhances academic management and learning through intelligent automation. Teachers can manage attendance, marks, and assignments, and upload materials that train a subject-specific AI Tutor to answer student queries based only on teacher-provided content.\r\nUsing a Progressive Unlocking System, course materials are released gradually (e.g., week-by-week) for structured and focused learning. An AI-powered Plagiarism Checker evaluates submissions and also compare inter-student assignments to identify content similarities and detect plagiarism.\r\nStudents benefit from smart notifications, deadline alerts, and assignment severity indicators that help them prioritize tasks. A performance scale dashboard visually tracks academic progress across each subject. \r\nAdditionally, the system provides insights into next semester subjects, including their detailed outlines and recommended prerequisite topics to learn beforehand, helping students prepare early and bridge knowledge gaps in advance.\r\nTogether, these features ensure a personalized, fair, and efficient academic experience for both teachers and students.","description":"The Smart University Mobile App is a Flutter-based cross-platform system designed for both Teachers and Students. Teachers can manage attendance, marks, timetables, and assignments, and upload course materials (PDFs, slides, notes) that train a subject-specific AI Tutor. This tutor provides answers based only on teacher-provided content and follows a Progressive Unlocking approach, where materials are released gradually (e.g., week-by-week) for organized learning .The system includes an AI-powered Plagiarism Checker that evaluates submissions and compares student assignments to detect similarities and ensure academic integrity. Students receive smart notifications, deadline alerts, and task submission deadline severity indicators to prioritize tasks efficiently. A performance scale dashboard helps track subject-wise progress with visual severity indicator. This will help students to prioritize subject learning. The University AI Bot assists with general campus queries .Additionally, students can view next semester subjects with details and recommended prerequisite topics to prepare before the course begins. Together, these features create a personalized, intelligent, and efficient academic ecosystem for both teachers and students$$\nAuthentication & Role Management\r\n\u2022 Provides secure login and logout for both Teachers and Students.\r\n\u2022 Ensures role-based access \u2014 Teachers manage academic modules, while Students access learning and progress features.\r\n\u2022 Developed using Firebase Authentication with session management to maintain data security, reliability, and a smooth user experience.\r\n\r\nTeacher Dashboard & Role Modules\r\n\u2022 Teacher Dashboard: A centralized interface allowing teachers to manage attendance, marks, timetables, content uploads, and student submissions efficiently.\r\n\u2022 Attendance Management: Enables teachers to mark and update attendance; the system automatically generates attendance summaries.\r\n\u2022 Marks & Assignment Upload: Teachers can upload, edit, and manage marks and assignments. This module integrates with the AI-Powered Plagiarism Checker for similarity analysis.\r\n\u2022 AI-Powered Plagiarism Checker: Evaluates student submissions to detect content similarities and plagiarism percentage, ensuring academic honesty by alerting both teachers and students.\r\n\u2022 Content Upload: Teachers upload lecture slides, PDFs, notes, and books, stored securely in the backend and used to train a subject-specific AI Tutor.\r\n\u2022 Progressive Unlocking: Allows teachers to upload all content early while releasing it gradually (e.g., week-by-week). The AI Tutor learns from all materials but answers only from unlocked portions to maintain course structure.\r\n\u2022 Prerequisite Management: Teachers define prerequisite topics for upcoming courses, guiding students on what to study beforehand.\r\n\u2022 AI Course Tutor (Teacher Assistant): Each course has its own AI assistant trained on that teacher\u2019s content, offering Q&A, lecture summaries, and quiz generation.\r\n\r\nStudent Dashboard & Role Modules\r\n\u2022 Student Dashboard: A unified panel showing attendance, marks, assignments, timetables, prerequisites, and analytics in one place.\r\n\u2022 Assignment Submission with Auto-Check: Students upload assignments directly, automatically evaluated by the AI-Powered Plagiarism Checker before final submission.\r\n\u2022 Notifications & Reminders: Sends smart alerts for assignments, attendance, unlocked content, and university updates.\r\n\u2022 Assignment Severity & Prioritization: Displays severity (High, Medium, Low) based on weightage and deadline to help students prioritize.\r\n\u2022 Scholarship Section: Lists scholarships with eligibility details, deadlines, and application links.\r\n\u2022 Performance Scale Dashboard: Visually represents subject-wise performance to help students track academic progress.\r\n\u2022 AI Learning Tutor: A subject-specific AI assistant trained on teacher-provided materials, offering interactive Q&A, summaries, and quiz generation, restricted to unlocked content.\r\n\u2022 Prerequisite & Semester Planner: Displays next semester subjects with recommended prerequisites to aid preparation.\r\n\u2022 University AI Bot: A virtual assistant answering general campus queries like fee deadlines, registration, and academic calendar details.\r\n\r\nAI Engine Modules\r\n\u2022 Data Ingestion: Extracts and processes data from teacher-uploaded files (PDFs, slides, notes, and books).\r\n\u2022 Vector Database: Maintains secure, separate knowledge bases for each teacher\u2019s content to ensure privacy.\r\n\u2022 Retrieval-Augmented Generation (RAG): Ensures the AI Tutor retrieves and generates responses only from relevant, course-specific teacher data.\r\n\u2022 Q&A, Summarization, and Quiz Generation: Provides academic support by answering questions, summarizing lectures, and generating quizzes automatically.\r\n\u2022 Plagiarism Detection Engine: Uses AI-based similarity detection to analyze assignments, identify plagiarism, and highlight matching submissions.$$\n(i) Student Dashboard: Displays attendance, marks, assignments, timetables, prerequisites, and overall progress in one interface.\r\n(ii) Assignment Submission with Auto-Check: Lets students upload assignments automatically analyzed by the AI-Powered Plagiarism Detection Engine before final submission, ensuring originality.\r\n(iii) Scholarship Section: Shows available scholarships with eligibility criteria, deadlines, and direct links for easy application.\r\n(iv) Notifications & Reminders: Sends alerts for assignment deadlines, attendance shortages, new content, and scholarship updates to keep students informed.\r\n(v) AI Learning Tutor: A subject-specific AI assistant trained on teacher-provided materials. It delivers interactive Q&A, lecture summaries, and quiz generation restricted to unlocked content for focused learning.\r\n(vi) Prerequisite & Semester Planner: Suggests upcoming subjects and recommended prerequisite topics, helping students prepare ahead.\r\n(vii) University AI Bot: Virtual helpdesk responding to general university queries like registration details, fee deadlines, and academic updates.$$\n(i) Attendance Management: Enables teachers to mark, update, and track attendance efficiently. The system automatically generates attendance summaries for analytics.\r\n(ii) Marks & Assignments Upload: Lets teachers upload, edit, and manage marks, quizzes, and assignments. Integrated with the AI-Powered Plagiarism Checker to detect similarities and ensure integrity.\r\n(iii) Timetable Management: Provides tools to create, update, and share schedules in real time, keeping teachers and students informed.\r\n(iv) Content Upload: Enables uploading of slides, notes, and books, securely stored and used to train the course AI Tutor.\r\n(v) Progressive Unlocking: Allows scheduling of content visibility so materials release gradually. The AI Tutor answers queries only from unlocked portions, promoting structured learning.\r\n(vi) Prerequisite Management: Lets teachers define prerequisite topics or skills for upcoming subjects, helping students prepare early.\r\n(vii) AI Course Tutor: Creates a personalized AI Tutor trained on teacher materials. It provides accurate answers, summaries, and quizzes to support learning.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-08 16:34:30","updated_at":"2025-10-20 12:38:07","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":965,"project_id":1318,"title":"VissaAssist CRM","prob":"This FYP addresses inefficiencies in the visa application and immigration process, including manual lead management, complex document handling, and limited client support. Current systems often result in delays, errors, and poor user experience due to time-consuming processes and lack of automation. VissaAssist CRM tackles these issues by automating lead management, streamlining contract drafting, and integrating AI features like a 24\/7 chatbot and Gemini-powered tools. This enhances efficiency, reduces errors, and improves client satisfaction by providing real-time assistance and actionable insights through comprehensive reporting.","description":"VissaAssist CRM is a comprehensive customer relationship management system designed to optimize the visa application and immigration process. It manages client lead generation, sales processing, operational workflows, financial tracking, and administrative tasks. By integrating AI features such as an intelligent chatbot and Gemini-powered drafting, the system enhances client interactions and automates complex document creation, delivering an efficient and cutting edge visa service platform.\r\nPrimary Objectives:\r\nSimplify visa program browsing and application submission.\r\nAutomate lead management and client engagement.\r\nStreamline contract drafting and operational workflows.\r\nEnhance client experience with reports and 24\/7 support.\r\nProvide financial and administrative controls.$$\nClient Module: Browse and filter visa and immigration programs by various criteria; submit detailed application forms online; receive notifications and track application status.\r\nSales Module: Manage incoming leads from application submissions; contact clients for program confirmation and document collection; schedule follow-ups and monitor lead progress; log client communications and document validation.\r\nOperation Module: Support contract creation with one time or installment payments; manage document verification, submission tracking, and immigration authority follow-ups and deadlines.\r\nFinance Module: Generate and dispatch client invoices; track payment receipts and financial transactions; manage internal office expenses including payroll; produce financial reports like cash flow and outstanding payments.\r\nAdmin Module: Handle user and role management with granular permission controls; configure system and security settings; monitor activities and maintain audit logging; provide cross-module analytical and operational reporting.\r\n\r\nAI Features (Phase 1):\r\n\r\nAI Chatbot for Client Support: Offers a 24\/7 conversational assistant on the client portal, provides instant FAQ answers and visa guidance.\r\n\r\nGemini-Powered Contract Drafting & Program Descriptions: Automatically generates tailored visa contracts based on client and program data; produces clear, engaging program descriptions from structured inputs; reduces manual effort and improves consistency.$$\nI will develop the Client Module (program browsing, application submission, status tracking), Sales Module (lead management, client communications, follow-ups), and RBAC and Authentication Module (JWT-based access control and authentication). I will also integrate the AI Chatbot for 24\/7 support.$$\nI will develop the Operation Module (contract creation, document verification, authority follow-ups), Finance Module (invoicing, payment tracking, expense management, financial reports), and Admin Module (user management, security, analytical reporting). I will integrate the Gemini API for contract drafting and program descriptions.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Your idea currently involves several manual processes. Consider incorporating automation for information sharing to improve efficiency and reduce human effort. $$ Student were advised to give a comparison with these FYPs (StudyHub-An study consultancy Application,CONSULTATION FOR YOU (Online Visa and Immigration Consultation for Students, Families, and Tourists) which are already available on RMS. Also it is suggested to make it a market place for the consultancy agencies.","isDraft":0,"status":2,"created_at":"2025-09-30 19:42:32","updated_at":"2025-10-22 12:18:15","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1047,"project_id":1336,"title":"DeceptiVerse: AI-Powered Dynamic Deception Framework Using Fake Web Applications and Deepfake Audio","prob":"This project solves the problem that traditional honeypots are static and predictable, making them easy for attackers to identify and bypass. This leaves organizations without detailed behavioral data on their attackers.\r\n\r\nDeceptiVerse solves this by using AI-driven deception to create \"fake yet realistic web environments\". It dynamically builds fake dashboards and \"uses deepfake\" audio to \"mislead, detect, and study attacker behavior in real-time\". This approach helps organizations gain \"rich behavioral intelligence\" and \"protect real assets\".","description":"DeceptiVerse is an advanced AI-based deception framework designed to mislead, detect, and study cyber-attackers. It addresses the core problem of traditional honeypots, which are static, predictable, and easily bypassed by attackers.\r\n\r\nThe system works by dynamically generating a unique, realistic fake web environment\u2014such as an HR, Finance, or CRM dashboard\u2014for each attacker session. This dynamic nature is enhanced by an \"Adaptive Environment Engine\" and \"Deepfake Audio Deception,\" which uses fake voice prompts to make the trap more believable and engaging.\r\n\r\nHere is the working flow:\r\n1. An attacker reaches the fake portal.\r\n2. DeceptiVerse instantly generates a unique session and fake dashboard for them.\r\n3. An \"Interaction Logger\" records every action the attacker takes (e.g., clicks, logins, uploads), automatically building a chronological behavioral timeline.\r\n4. This timeline is fed in real-time to the \"Attacker Behavior Analyzer,\" an AI module using machine learning (like TensorFlow). This \"brain\" analyzes the behavior, detects suspicious patterns, classifies the attacker's skill or intent, and assigns a risk level.\r\n5. If high-risk behavior is detected, the \"Alert Manager\" automatically sends real-time alerts to an organization's security (SOC\/MDR) dashboards for immediate response.\r\n\r\nThe project's goal is to provide organizations with rich, actionable intelligence on attacker behavior, helping to reduce detection time and enhance overall cyber defense.$$\n1. Fake Web App Generator\r\nThis module is the core of the deception and actrecognizingrchestrator\" for the entire framework. Its primary function is the \"AI-driven dynamic fake web app generation per attacker session\"\r\n\r\nInstead of a single, static honeypot that can be easily identified, this module dynamically builds and deploys realistic-looking fake web applications (such as HR, CRM, or Finance dashboards) using React. It ensures that \"each attacker session receives a unique environment with fake data\" , such as different employee names, sales figures, or project details. This makes the deception unique to each attacker, preventing them from recognizing a predictable pattern.\r\n2. Attacker Behavior Analyzer behavioral the \"brain\" of the project, responsible for generating actionable intelligence. This module performs two main tasks: logging and analysis.\r\n\r\nLogging: It acts as an \"Interaction Logger\" that \"captures each attacker\u2019s actions\" , such as every login attempt, click, form submission, and file upload. This data is used to \"build a behavioral timeline\" [cite: 2, file 1] which is stored in the MySQL database.\r\n\r\nAnalysis: It then \"applies AI models to detect patterns\" using TensorFlow. The goal is to \"analyze attacker behavior and classify skill level or intent\", assigning a risk level to their actions. This module provides the \"behavioral profiling\" that powers the entire system.\r\n3. Deepfake Audio Deception\r\nThis is an innovative module designed to significantly enhance the realism of the fake environments. It directly addresses a key gap in traditional systems, which have \"No Voice Deception\u201d.\r\n\r\nThis module uses the ElevenLabs API to generate realistic, AI-driven deepfake audio. These audio files are then \"used to engage intruders\" by being embedded within the fake web apps. For example, an attacker might find what appear to be recorded voicemail messages or clips from meetings. These \"voice prompts\"] make the fake environment far more immersive and believable, encouraging the attacker to stay engaged for longer.\r\n4. Adaptive Environment Engine\r\nThis module connects all the other components to make the deception truly \"smart\" and \"adaptive\". It is responsible for the project's \"adaptive deception logic\".\r\n\r\nThis engine takes the real-time \"behavioral profiling\" data from the \"Attacker Behavior Analyzer\" (Module 2) and uses it to automatically change the \"Fake Web App\" (Module 1). For example, if the analyzer determines an attacker is searching for financial data, this engine will \"automate creation of adaptive deception environments based on attacker profile\". It might instantly create and display a new, fake \"Quarterly_Report.pdf\" file for the attacker to find, guiding their behavior and misleading them further.$$\nMustajab will develop the modules responsible for creating and enhancing the deceptive environment:\r\n\r\n1. Fake Web App Generator: This is the primary module for this member. It involves building the dynamic, realistic fake web applications (HR, CRM dashboards) using React. This member will be responsible for the entire frontend \"trap.\"\r\n\r\n2. Deepfake Audio Deception: This module logically supports the \"Fake Web App Generator.\" It involves using the ElevenLabs API to generate fake audio files and embedding them into the fake dashboards to make the deception more believable.$$\nUmer Lodhi will develop the \"brain\" of the project, focusing on data analysis and adaptive logic:\r\n\r\nAttacker Behavior Analyzer: This is the primary module for this member. It involves building the AI models using TensorFlow to analyze the log data from MySQL, classify attacker behavior, and detect patterns.\r\n\r\nAdaptive Environment Engine: This module is the advanced logic that connects to the \"Analyzer.\" It will use the \"attacker profile\" to make real-time decisions and change the fake environment, making the deception \"adaptive\".$$\n$$\nnone$$\nAI-driven dynamic fake web app generation per attacker session$$\nBehavioral profiling and adaptive deception logic using AI$$\nDeepfake Audio Deception","comments":" $$ Approved with current scope","isDraft":0,"status":2,"created_at":"2025-11-06 00:14:30","updated_at":"2025-11-12 12:45:16","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":996,"project_id":1339,"title":"Ustad Go: AI-powered platform for hiring and managing verified local service providers.","prob":"In Pakistan, finding reliable and verified local service providers is a major challenge. Most people rely on personal references, phone numbers, or unverified online ads when they need services like electricians, plumbers, carpenters, masons, tutors, or domestic helpers. This process is often time-consuming, inconvenient, and unsafe, as there is no guarantee of trust, quality, or accountability. Lack of proper verification, secure communication, and standardized feedback makes users hesitant to book services digitally. The absence of a structured, accessible, and trustworthy platform creates uncertainty for both service seekers and providers, limiting opportunities and causing frustration in everyday service needs.","description":"Ustad Go is an AI-powered mobile application that allows users to book local services on demand such as electricians, plumbers, tutors, and domestic helpers. Like booking a ride, users can search by category, location, price, and availability to quickly access trusted providers. The platform ensures safety through CNIC verification, secure demo payments, and user logs, while an AI-powered recommendation system with priority handling suggests the best providers based on urgency, ratings, and availability.\r\nFor communication, Ustad Go supports in-app chat with text, voice, transcription, and Urdu language, ensuring inclusivity. Providers can manage profiles, availability, and bookings, while seekers benefit from dynamic ratings, feedback, and suggestions. A web-based admin dashboard manages verification, disputes, payments, and analytics. The app also enables AI issue detection from pictures with automated solutions, making service booking faster, safer, and more intelligent.$$\nMain Modules of Ustad Go\r\n\r\n1. Service Provider Module\r\nThis module enables local service providers (e.g., plumbers, electricians, tutors) to register, create professional profiles, and manage their availability. Providers can update skills, experience, and service charges while ensuring their identities are verified for user trust.\r\n\r\n2. Service Seeker Module\r\nThe service seeker module is designed for customers who want to hire local professionals. It allows them to browse providers, check profiles, compare ratings, and initiate service requests easily through the app.\r\n\r\n3. Booking & Scheduling Module\r\nThis module handles service booking and scheduling. Users can book providers for immediate service or set appointments for future dates. Both parties receive confirmations and reminders, ensuring a smooth workflow.\r\n\r\n4. AI Issue Detection & Solution Module\r\nAn AI-powered feature that allows service seekers to describe or upload details (text, images, or voice). The AI analyzes the problem and suggests potential issues and solutions, helping match the right service provider quickly.\r\n\r\n5. AI Recommendation System Module\r\nThis module leverages AI algorithms to recommend the most suitable service providers based on seeker requirements, location, provider ratings, and past preferences. Additionally, the system prioritizes recommendations by considering urgency of the request, real-time availability of providers, and past service history, ensuring faster and more reliable matches.\r\n\r\n6. Communication Module\r\nThis module facilitates interaction between seekers and providers. It includes:\r\n\u2022\tIn-App Chat for quick text communication.\r\n\u2022\tVoice Support for direct discussions.\r\n\u2022\tTranscription for converting voice to text.\r\n\u2022\tUrdu Language Support to enhance accessibility for local users.\r\n\r\n7. Notifications & Alerts Module\r\nReal-time push notifications and alerts keep users informed about booking status, service provider arrival, payments, promotions, and urgent updates.\r\n\r\n8. Ratings & Feedback Module\r\nAfter a service is completed, seekers can rate providers and leave feedback. This helps maintain transparency, build trust, and improve the platform\u2019s service quality over time.\r\n\r\n9. Admin Panel Module\r\nA backend control system for administrators to manage users, monitor activities, verify providers, resolve disputes, and analyze platform performance. It ensures security and smooth operations of the ecosystem.\r\n\r\n10. Search & Geolocation Module\r\nThis module allows seekers to search for nearby service providers using keywords, categories, or filters. Integrated geolocation ensures accurate distance calculation, estimated arrival times, and route guidance.\r\n\r\n11. Demo Payment Module\r\nFor demonstration purposes, this module simulates digital payment integration. It shows how users could securely pay providers via the app using wallets, cards, or local gateways in future versions.$$\n1.\tService Provider Module \u2013 self-registration with CNIC verification, profile setup, service listings, availability calendar.\r\n2.\tBooking & Scheduling (Provider\u2019s side) \u2013 receiving requests, accepting\/rejecting\/countering offers, managing confirmed bookings.\r\n3.\tAI Issue Detection & Solution Module \u2013 analysing uploaded pictures, detecting problems, and generating AI-based solutions.\r\n4.\tNotifications & Alerts Module \u2013 real-time updates for confirmations, reminders, and service changes.\r\n5.\tAdmin Panel \u2013 CNIC verification, dispute handling, provider monitoring.\r\n6.\tSearch & Geolocation Module \u2013 discovering services by location, map-based search.$$\n7.\tService Seeker Module \u2013 registration, profile creation, search, and viewing booking\/payment\/chat history.\r\n8.\tBooking & Scheduling (Seeker\u2019s side) \u2013 sending requests, proposing\/negotiating offers, managing booking history, invoices.\r\n9.\tAI Recommendation System Module \u2013 personalized and priority-based provider suggestions.\r\n10.\tCommunication Module \u2013 in-app chat with text, voice, transcription, and Urdu language support.\r\n11.\tRatings & Feedback Module \u2013 dynamic ratings, reviews, service feedback, and app-level feedback.\r\n12.\tDemo Payment Module \u2013 secure demo payment gateway integration for transactions.$$\n$$\nOnline Service Provider (OSF)$$\nAI Issue Detection & Solution: Users can upload a picture of a problem, and the system detects the issue and generates possible solutions using generative AI.$$\nAI-Powered Recommendation with Priority Handling: The system recommends providers not only based on ratings and popularity but also urgency, availability, and past service history.$$\nIn-App Multi-Mode Communication: Communication upgraded with voice chat, transcription, and Urdu language support, beyond traditional text-only chat.","comments":"","isDraft":0,"status":2,"created_at":"2025-10-08 10:16:27","updated_at":"2025-10-08 12:48:49","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1010,"project_id":1322,"title":"AI-Powered Adaptive Learning Platform with Personalized Pathways","prob":"Traditional e-learning systems follow a fixed approach that fails to adapt to individual learning styles, pace, and knowledge levels. This leads to student disengagement, poor retention, and ineffective outcomes. Educators also face challenges in creating quality content, assessing performance, and providing timely feedback to large groups. The proposed FYP solves these problems by introducing an AI-powered adaptive learning platform that personalizes education for every learner. It dynamically adjusts quiz difficulty, recommends relevant courses, and identifies a student\u2019s preferred learning style using AI and ML. The system also allows educators to auto-generate structured courses using the Gemini API and integrate global educational content from MIT, Coursera, and Khan Academy. This intelligent automation ensures a tailored, engaging, and scalable learning experience for both students and educators.","description":"The AI-Powered Adaptive Learning Platform with Personalized Pathways is a web-based application built using the MERN stack integrated with Python Flask for AI and Machine Learning services. It provides a smart, scalable environment where students and educators interact through adaptive, data-driven education.\r\n\r\nStudents can register via Google OAuth or email, explore courses, and enroll in either:\r\n\r\nEducator-created courses, built using an AI-powered course generation form integrated with the Gemini API.\r\n\r\nExternally fetched courses, automatically retrieved from APIs of MIT OpenCourseWare, Coursera, and Khan Academy.\r\n\r\nThe integrated Flask ML engine continuously analyzes each learner\u2019s quiz scores, progress, and behavior to adjust quiz difficulty, recommend suitable next courses, and predict performance trends. Educators can generate quizzes automatically using AI, manage course content, track student performance, and communicate in real time via chat.\r\n\r\nThe system features an AI Assistant (Vapi.ai) that provides voice and text tutoring, an adaptive quiz generator, automated certificate generation, and emotion-aware personalization to enhance engagement. The result is a highly intelligent and interactive learning ecosystem that bridges the gap between traditional learning and modern AI-driven education.$$\n(i) Authentication & Authorization Module:\r\nHandles secure login, registration, and role selection (Student\/Educator) using JWT and Google OAuth. Ensures secure access through encrypted tokens and role-based route protection.\r\n\r\n(ii) Student Dashboard:\r\nDisplays enrolled courses, learning progress, AI-based recommendations, and quiz performance analytics. Provides access to AI chat assistant, certificates, and real-time chat.\r\n\r\n(iii) Educator Dashboard:\r\nAllows educators to upload, edit, and manage courses, monitor student activity, generate quizzes, and access detailed analytics reports.\r\n\r\n(iv) AI Course Creation Module (Gemini Integration):\r\nEmpowers educators to auto-generate structured courses by entering course details. The Gemini API produces a full course layout, content, and images dynamically, saved into MongoDB.\r\n\r\n(v) External Course Integration Module:\r\nAutomatically fetches course data from MIT OpenCourseWare, Khan Academy, and Coursera APIs, enabling hybrid access to both in-house and global content.\r\n\r\n(vi) Flask ML Engine (Personalization):\r\nAnalyzes student data to detect learning style, predict success, adjust quiz difficulty, and recommend personalized courses. Continuously improves using user performance trends.\r\n\r\n(vii) AI Quiz Generator & Adaptive Assessment Module:\r\nUses Gemini or OpenAI APIs to create quizzes automatically. Flask ML dynamically adjusts question difficulty and topics based on prior performance. Results and analytics are stored in MongoDB.\r\n\r\n(viii) AI Assistant (Vapi.ai):\r\nA multimodal voice and text tutor that provides instant academic help, explanations, and concept summaries using integrated AI models (Vapi\/OpenAI).\r\n\r\n(ix) Real-Time Chat Module:\r\nImplements instant communication between students and educators using Socket.io. Enables discussion, support, and live doubt resolution.\r\n\r\n(x) Certificate Generator:\r\nAutomatically issues downloadable PDF certificates upon meeting course completion criteria.\r\n\r\n(xi) Analytics & Reporting Module:\r\nCollects and visualizes learner performance data\u2014progress graphs, average scores, engagement rates, and predictive insights for educators.\r\n\r\nTogether, these modules form an intelligent, flexible, and scalable learning ecosystem driven by hybrid AI and ML technologies.$$\nI will develop the frontend and core system modules using the MERN stack, focusing on user experience and system functionality. My responsibilities include:\r\n\r\nDesigning and implementing the frontend UI using React.js, Tailwind CSS, and shadcn\/ui, ensuring responsiveness and usability.\r\n\r\nDeveloping the Authentication & Authorization Module (JWT, Google OAuth) and Role-Based Dashboards for Students and Educators.\r\n\r\nImplementing Course Management features (CRUD operations, enrollments, progress tracking).\r\n\r\nBuilding the Real-Time Chat Module using Socket.io and integrating the AI Assistant (Vapi.ai) for text and voice interaction.\r\n\r\nCreating the Certificate Generation System and connecting frontend logic to the Flask ML and AI APIs.\r\nMy goal is to deliver a seamless, user-friendly interface and ensure all backend AI services are efficiently integrated and accessible through the web application.$$\nI will develop the AI and ML integration modules, focusing on:\r\n\r\nThe Flask-based Machine Learning Engine for adaptive learning and personalized course recommendations.\r\n\r\nThe AI Quiz Generator and its backend logic to generate adaptive quizzes and predict difficulty using the Gemini and OpenAI APIs.\r\n\r\nThe Recommendation System, which analyzes quiz results and learning behavior to suggest suitable next courses.\r\n\r\nAPI integration with MIT OpenCourseWare, Khan Academy, and Coursera for external course fetching and mapping with the student\u2019s learning profile.\r\nMy work ensures that the platform\u2019s AI and ML components function cohesively, providing real-time personalization, accurate predictions, and adaptive course delivery.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-08 18:31:09","updated_at":"2025-10-09 11:14:47","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":998,"project_id":1359,"title":"EduBite\r\nA Smart Digital Marketplace for School\/College & University Canteens","prob":"Currently, school, college and university canteens operate manually with long queues, delayed services, and no intelligent way to plan or manage meals. Food is often wasted due to poor inventory control, and students\/faculty face difficulty in managing orders with their schedules. Existing food delivery apps are not designed for educational institutions and charge heavy commissions.\r\nOur system solves this problem by providing a digital marketplace platform exclusively for schools\/colleges and universities. Students and faculty can order meals from their registered canteens, plan weekly meals, and receive personalized recommendations. Canteen owners get a smart dashboard with inventory tracking and AI-based demand prediction to reduce wastage. This makes the system efficient, sustainable, and tailored for educational needs.","description":"This project will function as a multi-canteen digital marketplace, primarily developed as a web application with optional mobile access for users. Users can log in, see the daily menu, plan their meals for the week, and place orders. Payments will be made securely through integrated e-wallets (JazzCash, Easypaisa, Stripe) and users will get notifications when their food is ready. Canteen owners can manage their menus, monitor inventory, and analyze sales using smart dashboards. The system will include AI-based recommendations and demand prediction to ensure better meal planning and reduced wastage. Admins can oversee multiple canteens across different campuses through a centralized console.$$\n1.\tMulti-Canteen Marketplace\r\n\u2022\tEvery school\/college\/university canteen has its own profile, menu, and reports.\r\n\u2022\tStudents\/faculty can only see their own institute\u2019s registered canteens.\r\n2.\tUser Consoles (Role-Based Access)\r\n i) Canteen Owner Console \u2192 Menu upload, order management, stock control.\r\n ii) Student Console \u2192 Digital menu, instant & pre-orders, feedback.\r\n iii) Faculty Console \u2192 Digital menu, instant & pre-orders, feedback.\r\n iv) Admin Console (Super Admin) \u2192 Approves new canteens, monitors analytics.\r\n3.\tAI Modules\r\n i) AI Demand Prediction \u2192 LSTM for dish-level demand forecasting.\r\n ii) AI Recommendations \u2192 Collaborative + Content-based filtering.\r\n4.\tWeekly Meal Planning & Pre-Orders\r\n\u2022\tStudents\/Teachers set meal plans in advance.\r\n\u2022\tAuto-confirm daily orders via scheduler.\r\n\u2022\tCancelation allowed before cutoff time.\r\n5.\tPayments & E-Wallet\r\n\u2022\tJazzCash\/EasyPaisa\/Stripe API integration.\r\n\u2022\tBuilt-in wallet for instant transactions.\r\n6.\tSmart Inventory & Wastage Analytics\r\n\u2022\tDashboard compares sales vs stock.\r\n\u2022\tPredicts raw material needs.\r\n\u2022\tAlerts for shortage\/wastage.\r\n7.\tLive Chat with Staff\r\n\u2022\tReal-time messaging between students & canteen staff.\r\n\u2022\tFirebase + MongoDB chat logs.\r\n8.\tFeedback System\r\nUsers can rate meals and give feedback to improve service quality.$$\nI will develop the Canteen Owner and Backend Modules with Node.js and Firebase, featuring secure login, payment gateway integration, intuitive admin dashboard, and real-time live chat support for seamless customer communication and efficient order\u00a0management.$$\nI will develop the Student Module with React.js, leveraging TailwindCSS\/Bootstrap for a responsive UI, featuring weekly meal planning, personalized recommendations, and a feedback system for an enhanced user experience and\u00a0engagement.$$\n$$\n\u2022\tSCMS (Smart Cafe Management System)\r\n\u2022\tCOMSAT\u2019s Cafe System$$\n1. Multi-Canteen Marketplace \u2013 a centralized platform for schools, colleges and universities.$$\n2. AI-Powered Features \u2013 Personalized recommendations + demand forecasting.$$\n3. Comprehensive Modules \u2013 Separate consoles for students, faculty, canteen owners, and admins.","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-08 11:57:01","updated_at":"2025-10-20 12:40:35","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1045,"project_id":1398,"title":"AgriVision: AI Diagnosis and Orchard Trade Hub","prob":"Fruit orchard owners in Pakistan face two major challenges: (1) late or incorrect disease identification (Black Spot, Citrus Canker, Greening, Fruit Rot), causing large yield and quality losses; and (2) weak market connectivity, farmers rely on middlemen and lack transparent channels to buyers\/exporters, with no reliable payment or harvest tracking. AgriVision solves both by combining AI-based fruit disease detection and spray recommendations with a mobile marketplace for direct, transparent farmer\u2013buyer deals.","description":"AgriVision is an AI-powered mobile-based platform that assists fruit orchard owners in disease detection and direct market connectivity. Farmers capture leaf\/fruit photos; images are preprocessed (resize, normalize, augment) and sent to a CNN (EfficientNet-B0) for classification (Healthy, Canker, Black Spot, Greening) with confidence and severity estimates. The app displays context-aware spray recommendations drawn from a curated knowledge base and adjusted for weather and prior spray history. Through the AgriVision app, farmers can create profiles, register orchards, and upload fruit or leaf images for AI-based disease detection and spray recommendations. They can also list harvests with photos, quantity, and price on the marketplace. Buyers can create profile, browse listings, filter by fruit type or price, and contact farmers via built-in call to finalize deal.$$\n1. Data Collection & Preprocessing Module: This module manages the acquisition of fruit and leaf imagery from Kaggle datasets, supplemented by manually collected samples from local orchards to enhance dataset diversity. All farmer-uploaded images undergo standardized preprocessing, including resizing, normalization, noise reduction, and augmentation techniques such as rotation, brightness adjustment, and zoom variation. Metadata including orchard identification, GPS coordinates, and timestamps are systematically stored to facilitate continuous model retraining and performance optimization. 2. Disease Detection & Classification Module (CNN-Based): The core AI component employs a Convolutional Neural Network architecture (EfficientNet-B0) to identify and classify plant diseases in real-time. Upon image upload via the mobile application, the system analyzes the visual data and detects conditions including Citrus Canker, Black Spot, Greening Disease, and Fruit Rot. Each prediction includes a disease classification label and confidence score, which are archived for historical analysis and integrated into the continuous learning pipeline to improve detection accuracy over time.\r\n3. Spray Recommendation Engine: This module generates automated treatment and spray recommendations based on the detected disease and external weather API data. It provides dosage details, application intervals. The recommendation engine cross-verifies the orchard\u2019s spray history to prevent overuse and ensures optimal scheduling, improving yield and reducing environmental harm. 4. Harvest Stock Listing (Farmer Side): This module enables farmers to create and manage detailed harvest listings for their orchards. Each listing includes essential information such as fruit variety, quantity, quality grade, expected price, and harvest date. Farmers can also upload images of fruit to determine fruit quality. Once submitted, listings become visible on the marketplace, allowing buyers to explore available stock. The module automatically updates listing statuses as transactions progress, helping farmers track which batches are sold, pending, or ready for harvest. 5. Marketplace & Search (Buyer Side): This module serves as the digital marketplace where buyers can explore, filter, and compare orchards or harvest listings. Buyers can search based on parameters such as fruit type, variety, price range, quality grade, and harvest date. Each orchard or harvest listing displays comprehensive details, including the disease history detected by the AI system, images, and past yield performance. Buyers can initiate negotiations. The system ensures transparency by presenting verified data, helping buyers make informed purchase decisions and fostering trust between both parties. 6. Order & Deal Management: This module handles the complete lifecycle of an orchard transaction \u2014 from the initial buyer request to final delivery and payment. Each order progresses through predefined stages: Requested \u2192 Confirmed \u2192 Advance Paid \u2192 Harvest \u2192 Delivered \u2192 Final Payment. Farmers and buyers can view live order statuses and track updates in real time. Payment records (advance and final) are maintained for accountability, even though actual transactions occur offline. The module also ensures automatic status updates of orchard listings (e.g., from Available to Sold once the deal is completed), maintaining marketplace accuracy.$$\nIn AgriVision, I shall develop the AI and Management modules which include: 2. Disease Detection & Classification (CNN-Based) 4. Harvest Stock Listing (Farmer Side), 6. Order & Deal Management.$$\nIn AgriVision, I shall develop the Data and Visualization modules which include: 1. Data Collection & Preprocessing, 3. Spray Recommendation Engine, \r\n5. Marketplace & Search (Buyer Side)$$\n$$\n1. AGRAZ ( A MARKETPLACE FOR FARMERS ) \r\n2. Plantish: An AI application for plant disease detection using deep learning\r\n3. Plant_Disease Detection$$\n1. AI-based Fruit Disease Detection System$$\n2. Smart Spray Recommendation Engine$$\n3. Integrated Marketplace & Deal Management","comments":"","isDraft":0,"status":2,"created_at":"2025-10-31 17:43:09","updated_at":"2025-11-05 09:10:55","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":990,"project_id":1385,"title":"PakPulse AI: Multi-Disease Outbreak Forecasting & Early Warning System for Pakistan","prob":"Pakistan suffers from recurring seasonal outbreaks of dengue, malaria, cholera, influenza, and COVID-19. Current systems are reactive, responding only after hospitals report rising cases, which leads to delayed interventions, hospital overcrowding, misallocation of resources, and preventable loss of lives. Existing solutions are limited, focusing on a single disease (e.g., dengue dashboards) and lacking predictive, real-time, and explainable capabilities. There is no unified system capable of forecasting multiple diseases simultaneously while providing transparency into how predictions are made. This FYP addresses this gap by building an AI-powered, multi-disease outbreak early warning system. It integrates real-time data from WHO, weather APIs, and simulated hospital feeds, applies deep learning models for 2\u20134 weeks advance forecasting, and generates a Composite Epidemic Risk Score (CERS). The system ensures explainability using SHAP and LIME and empowers policymakers through GIS dashboards and automated alerts.","description":"PakPulse AI is a web-based AI-driven health surveillance system designed to forecast and detect outbreaks of multiple diseases in Pakistan, including dengue, malaria, cholera, influenza, and COVID-19. The system integrates real-time climate data from OpenWeatherMap API, official disease statistics from WHO GHO and Athena APIs, and synthetic hospital feeds that simulate local case reporting.\r\nThe backend applies machine learning classifiers (Random Forest, XGBoost) for outbreak risk detection and a deep learning module (LSTM\/GRU) for forecasting disease cases 2\u20134 weeks in advance. A novel feature, the Composite Epidemic Risk Score (CERS), combines environmental, population, sanitation, and disease history factors into a single risk score per district.\r\nThe system ensures transparency using Explainable AI (SHAP & LIME) to justify why a prediction was made. The frontend includes a GIS-enabled dashboard developed in Streamlit, displaying interactive district-level risk heatmaps, time-series forecasts, CERS meters, and alerts. Automated email\/SMS notifications are sent when risk levels exceed defined thresholds.\r\nPakPulse AI aims to shift Pakistan\u2019s epidemic response from reactive crisis management to proactive prevention, enabling NIH, provincial health departments, and hospitals to prepare earlier, allocate resources effectively, and reduce preventable disease burden.$$\nData Collection & Storage Module\r\nThis module fetches health and climate data from WHO GHO\/Athena APIs, OpenWeatherMap, and synthetic hospital feeds. The collected data is stored in a DBMS (MySQL) and normalized into a time-series format for further processing.\r\n\r\nAI & Forecasting Module \r\nMachine learning classifiers (Random Forest, XGBoost) are applied for outbreak detection, while deep learning models (LSTM\/GRU) forecast disease trends 2\u20134 weeks ahead. Anomaly detection methods like Isolation Forest and DBSCAN ensure unusual patterns are identified early.\r\n\r\nComposite Epidemic Risk Score \r\nThis module generates a district-wise health risk score (0\u2013100) by integrating climate data, outbreak history, sanitation levels, and population density. The score acts as a simple yet comprehensive epidemic risk indicator.\r\n\r\nExplainable AI Module \r\nSHAP and LIME techniques provide interpretability of predictions. SHAP explains global feature importance (e.g., rainfall driving dengue risk), while LIME gives case-specific explanations for individual forecasts.\r\n\r\nDashboard & GIS Visualization Module\r\nAn interactive Streamlit-based dashboard presents outbreak risks with GIS maps (Folium\/Plotly), time-series forecasts, and district-level epidemic risk meters (CERS). This allows decision-makers to visualize and compare risks clearly.\r\n\r\nAlert & Notification Module \r\nThreshold-based alerts are generated when potential outbreaks are detected. Notifications are sent via email (SMTP) and SMS (Twilio API), with alert logs stored in the DBMS for record-keeping and review.\r\n\r\nUser Management & Authentication Module\r\nProvides secure access control with login\/logout functionality and role-based permissions for administrators and health officers to ensure safe system usage.$$\nI will develop the AI & Forecasting pipeline, which includes building ML classifiers (Random Forest, XGBoost), training deep learning LSTM\/GRU models for outbreak forecasting, generating Composite Epidemic Risk Scores (CERS), and applying Explainable AI (SHAP\/LIME) for model transparency. My work ensures that the system has strong predictive and AI depth.$$\nI will develop the UI\/UX, dashboard, and alerting modules, including the GIS-enabled dashboard in Streamlit, visualization of district-level risks and LSTM forecasts, integration of CERS meters, and automated email\/SMS alert system. My focus is to make the system user-friendly, visually interactive, and actionable for policymakers.$$\n$$\nThere is no similar kind of project on RMS$$\nAn interactive Streamlit web application where users (e.g. health officers) can view district-level outbreak maps, forecast graphs, risk scores (CERS), and SHAP-based explanations.$$\nConverts forecast results into practical, rule-based recommendations (e.g fogging, awareness drives, hospital alerts). This ensures the system moves from data prediction to actionable decision making$$\nSends real-time email\/SMS notifications when risks exceed thresholds and generates weekly PDF reports summarizing high-risk districts, forecasts, and suggested actions.","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-07 20:44:16","updated_at":"2025-10-20 12:42:36","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":962,"project_id":1320,"title":"CleanCore \u2013 AI-Driven Waste Monitoring and Optimization","prob":"Traditional waste management systems rely heavily on physical sensors in bins to monitor fill levels. While effective, this approach is costly and difficult to scale, especially across large urban areas with thousands of bins. Many municipalities struggle with high infrastructure costs, inefficient collection routes, and landfill overflow. CleanCore solves this problem by introducing a hybrid AI-driven waste monitoring platform that reduces dependency on hardware sensors by predicting bin fill levels and waste types using AI. This minimizes costs while ensuring accurate monitoring. The solution optimizes collection routes, reduces fuel consumption, and supports sustainability goals. It addresses real-world inefficiencies by enabling smarter, cheaper, and more environmentally friendly waste management practices.","description":"CleanCore is an AI-driven hybrid waste monitoring system developed in collaboration with SmartEnds. The platform integrates minimal sensor deployment with AI-based predictions to monitor bin fill levels and waste types across urban areas. The AI module predicts fill levels for non-sensor bins using historical data, image recognition, and reference sensor inputs. Confidence scoring and anomaly detection ensure reliability, triggering alerts where sensor deployment is necessary. Route optimization algorithms combine real-time and predicted data to generate efficient, eco-friendly collection schedules. The system includes dashboards and a mobile interface for stakeholders, enabling monitoring of operational efficiency, sustainability metrics, and financial performance. CleanCore addresses the cost barrier of sensor-heavy systems, positioning itself as a scalable, smart city-ready solution.$$\n1. Prediction & Monitoring Module\r\n2. Waste type classification using AI \r\n3. Confidence scoring and anomaly detection\r\n4. Route Optimization\r\n4. Data integration from minimal sensors\r\n5. User & Admin Modules (Dashboard and Mobile App)\r\n6. Data visualization Interface with Power BI\/Tableau\r\n7. Map-based interface and alerts system$$\nPrediction & Monitoring Module (Fill level prediction using image data)\r\nWaste type classification using AI (Object detection using TensorFlow\/PyTorch)\r\nConfidence scoring and anomaly detection \r\nData integration from minimal sensors$$\nUser Module (Mobile App Dashboard)\r\nAdmin Modules (Web- based)\r\nData visualization Interface with Power BI\/Tableau ( Data science module)\r\nMap-based interface and alerts system (Mobile App module)\r\nRoute Optimization$$\n$$\n1. Scrap App \u2013 A Mobile Recycling and Scrap Collection App\r\n2. Smart Waste Bin \u2013 IoT based fill-level monitoring \r\n\r\nYes, similar FYPs exist in recycling\/waste apps but they focus mainly on community scrap collection and awareness.$$\n1. Hybrid AI + Sensor System: Unlike purely sensor-based or AI-only systems, CleanCore combines both for reliability and cost reduction.$$\n2. Confidence Scoring \u2013 Each AI prediction will have a confidence score. Low-confidence results will be flagged for review.$$\n3. Anomaly Detection \u2013 Identifies unusual patterns (e.g., irregular fill rates, wrong waste type, faulty sensor data) and sends smart alerts","comments":"","isDraft":0,"status":2,"created_at":"2025-09-30 12:23:40","updated_at":"2025-10-08 12:41:26","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":991,"project_id":1349,"title":"ClarityWorks - Calibrated Reading Level & Smart Simplify","prob":"ClarityWorks solves the everyday mismatch between writing and readers. Schools, NGOs, support teams, and publishers need content that meets a defined level; but today they rely on guesswork, one-off scores, or black-box LLM rewrites with no proof. ClarityWorks provides a governed workflow: it measures difficulty against recognized bands (grade\/CEFR), pinpoints exactly why a passage is hard, and delivers controlled simplifications to a chosen target with a transparent before\/after view. It goes beyond simple text: OCR turns scans and photos into analyzable text, voice input captures spoken drafts, and a private reference library (RAG) grounds definitions and phrasing in your own textbooks, manuals, or style guides. Teams get repeatable, audit-ready results; exportable reports, saved history, role-aware defaults, and a lightweight admin panel, so teachers can match texts to classes, ESL programs adapt lessons, help centers deflect tickets, and publishers meet plain-language standards with evidence.","description":"ClarityWorks is a web application for assessing complexity and controlled simplification text. Users paste or upload a document (TXT, DOCX, PDF) and select Analyze. The system cleans the text, computes standard readability scores, and reports a single overall reading level (e.g., grade band), with clear explanations of difficult sentences (e.g., too long, uncommon vocabulary, unclear structure). If a target level is chosen, ClarityWorks proposes controlled rewrites that preserve meaning. A before\/after view and change summary make edits transparent.\r\nThe app supports scans and voice: OCR extracts text from scanned PDFs\/images, and speech input is transcribed and automatically structured for analysis and simplification. A private Reference Library (RAG) allows teachers or teams to upload textbooks, research papers, or glossaries; the system retrieves short, trusted snippets to define tricky terms and guide level-appropriate phrasing, with citations included in reports.\r\nResults are presented in a focused dashboard with a few simple visuals (overall level, sentence difficulty distribution, vocabulary mix). Users can export reports (PDF, DOCX, CSV\/JSON) capturing scores, flagged sentences, suggested edits, citations, and the before\/after diff. The platform includes sign-in, basic profiles, and a persistent history of analyses and rewrites. An admin panel provides lightweight user management and feature toggles.$$\n1.\tText Ingestion & Feature Extraction\r\nPaste text or upload common files (TXT, DOCX, PDF). Basic cleanup so the text is ready for analysis. Pull simple language signals (length, vocabulary variety, sentence structure). This module feeds both scoring and simplification.\r\n2.\tReadability Metrics\r\nShow well-known readability scores plus one overall level (e.g., CEFR\/Grade) based on analysis. Plain-language explanations of what each level means, who it is suitable for.\r\n3.\tDifficult-Sentence Detection & Explanations\r\nRanks hard sentences; shows short \u201cwhy\u201d flags for suggested changes (rare\/low-frequency words, complex sentences etc.) and quick edit tips.\r\n4.\tControlled Text Simplification\r\nUser selects target level for the text to be converted into; sentence-level suggestions with a before\/after difference of any changes in the text. Basic safeguards prevent oversimplification that could distort meaning\r\n5. Multimodal Capture: OCR & Voice Recognition\r\nOCR pipeline: Detects layout blocks in scanned PDFs and images, reads text, reorders by reading flow, and de-noises artifacts. Confidence scoring flags low-quality pages for quick manual review.\r\nVoice input (dictation): Users can speak drafts or notes; the system transcribes in real time, auto-segments into sentences, and immediately offers a \u201ctidy & structure\u201d pass so dictated text reads like well-formed prose. This makes it practical to capture lectures, field notes, and quick ideas and bring them into the same readability workflow.\r\n6. Retrieval-Augmented Knowledge: RAG Library\r\nLets teachers, editors, or teams upload textbooks, research papers, glossaries, or entire knowledge bases (single or batch). The system builds a private retrieval index (vectorized chunks with metadata). When a complex term is flagged, the system retrieves snippets to power on-hover definitions and to steer simplifications that preserve domain meaning. It also allows for gathering relevant information from multiple sources at one time for complexity analysis and simplifcation.\r\nConstrain rewrites: During simplification, retrieved passages are fed as context to the rewrite engine\/LLM so phrasing follows approved language and avoids hallucinations. Citations are attached in the report for auditability.\r\n7.\tDashboard & Visualization\r\nAll analysis results are presented in a clean dashboard that emphasizes clarity and accessibility. Key metrics (e.g., readability scores, sentence difficulty distribution) are visualized through intuitive charts. Plain language labels are used that are simple to understand.\r\n8.\tExports and Sharing\r\nUsers can export their analysis results and simplification outputs as structured reports. Export formats include standard documents (PDF, DOCX) for easy distribution. Reports capture metrics, flagged sentences, and simplification suggestions, ensuring that users can share insights or keep them for reference.\r\n9.\tUser Accounts and History\r\nThe platform supports user profiles with basic personalization options such as avatars. A persistent history is maintained: every text upload, analysis, and simplification is saved along with timestamps and version records. \r\n10.\tAdmin Panel\r\nFor administrators, the system includes a lightweight management interface. Admins can list and search user accounts, activate or deactivate users, reset passwords, and adjust global application settings. This ensures smooth operation and basic oversight without requiring deep technical expertise.$$\nJunaid will be responsible for developing advanced text analysis modules.\r\nHis work includes Text Ingestion & Feature Extraction, Readability Metrics, Controlled Text Simplification, and Retrieval-Augmented Knowledge: RAG Library. These modules focus on NLP-driven insights, readability evaluation, and smart simplification to improve clarity while preserving meaning.$$\nWahaj will develop core platform and user-facing features. His work covers Difficult-Sentence Detection & Explanations, Multimodal Capture: OCR & Voice Recognition, Dashboard & Visualization, Exports and Sharing, User Accounts, History & Collaboration, and the Admin Panel, ensuring smooth workflows, clear insights, and effective user and system management.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-07 21:48:58","updated_at":"2025-10-20 12:40:00","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1037,"project_id":1345,"title":"SchoolMate \u2013 Smart School Van Tracking & Management System","prob":"Parents often face challenges managing their children\u2019s daily school transportation due to a lack of route, driver, and timing transparency. They cannot track live locations or route deviations, and fee management is mostly manual, leading to confusion and miscommunication. Van owners also struggle to maintain student lists, attendance, and payments efficiently. SchoolMate solves these issues through a mobile-based, AI-powered platform connecting parents, van drivers, and schools. It offers real-time GPS tracking, AI-based ETA prediction, smart route optimization, and automated fee and attendance management. Parents receive live alerts for van arrival (within 300m), speed changes, or route deviations, while drivers manage routes and attendance through their app. Admins verify users, monitor vans, and ensure safety and transparency through dashboards and analytics.","description":"SchoolMate is a multi-stakeholder mobile app designed to make school transportation safe, transparent, and intelligent. It features three main interfaces \u2014 Parent App, Driver App, and Admin\/School Portal \u2014 each serving a specific purpose in the ecosystem. Parents can register their children, explore verified vans available nearby, assign vans based on route and timing, and track live locations with accurate AI-based ETA updates. Drivers can register themselves, manage van details, take attendance, and receive real-time alerts about route changes, delays, or student absences. The Admin\/School Portal enables verification of van owners and drivers, route approval, fee monitoring, and tracking of daily operations through an analytical dashboard. SchoolMate also includes features like route deviation detection, speed alerts, driver performance ratings, and multi-school and multi-route support, ensuring scalability, safety, and efficiency for all users.$$\n(i) Parent Module\r\n\u2022\tParent registration & login\r\n\u2022\tRegister children and assign vans\r\n\u2022\tView van, driver details, and live location tracking\r\n\u2022\tReceive AI-based ETA notifications\r\n\u2022\t\u201cVan is arriving\u201d alerts (within 300m)\r\n\u2022\tSpeed and route deviation alerts\r\n\u2022\tView child\u2019s attendance, trip history, and fee details\r\n\u2022\tGive feedback, rate drivers, and submit complaints\r\n(ii) Van Owner \/ Driver Module\r\n\u2022\tDriver registration & login\r\n\u2022\tAdd and manage van details, seats, and routes\r\n\u2022\tView assigned students and optimized pickup sequence\r\n\u2022\tMark daily attendance and update payment records\r\n\u2022\tLive route map navigation and alerts for delays\r\n\u2022\tSOS button for emergency reporting\r\n\u2022\tPush notifications to parents for trip start\/stop events\r\n(iii) Student History & Payment Module\r\n\u2022\tStore and manage student travel history\r\n\u2022\tTrack attendance and fee payments\r\n\u2022\tGenerate automated invoices and receipts\r\n\u2022\tSend reminders for pending payments\r\n\u2022\tGenerate reports and analytics for both parents and van owners\r\n(iv) Admin \/ School Portal\r\n\u2022\tVerify van owners, drivers, and documents\r\n\u2022\tManage parents, students, and vans\r\n\u2022\tApprove or block user accounts\r\n\u2022\tMonitor routes, vans, and payment activities\r\n\u2022\tDetect route deviations and unusual stops\r\n\u2022\tView reports, analytics, and system logs\r\n\u2022\tMulti-school and multi-route management$$\nPrimary Modules: Parent Module, Student History & Payment Module\r\nResponsibilities:\r\n\u2022\tDevelop parent and student interfaces\r\n\u2022\tImplement child registration, tracking, and AI ETA alerts\r\n\u2022\tManage payment tracking, invoices, and fee reports\r\n\u2022\tDesign complaint and feedback system$$\nPrimary Modules: Van Owner \/ Driver Module, Admin \/ School Portal\r\nResponsibilities:\r\n\u2022\tBuild driver-side app with route management and attendance tracking\r\n\u2022\tImplement admin verification and monitoring dashboard\r\n\u2022\tAdd route deviation alerts and driver rating functionality\r\n\u2022\tOptimize communication between driver and parent apps$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-10-21 19:52:33","updated_at":"2025-10-31 10:55:52","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1015,"project_id":1338,"title":"SmartWed 360: A Marketplace for Complete Wedding Management.","prob":"Planning a wedding is often stressful, time-consuming, and costly because families have to contact multiple vendors separately for venues, decor, catering, photography, transportation, and accommodation. This leads to miscommunication, delays, and confusion in organizing services. Existing platforms only provide simple vendor listings without smart search, price comparison, or intelligent suggestions.\r\nSmartWed 360 solves this problem by offering a centralized online marketplace where all wedding-related services are available in one place. It allows users to easily browse, compare, and book vendors based on their budget, location, and preferences. The system includes AI-based vendor recommendations, which help customers find the most suitable services quickly and efficiently.\r\nBy combining all essential wedding services under one platform, SmartWed 360 saves time, reduces planning stress, and provides a simple, transparent, and smart way for both customers and vendors to manage wedding arrangements more effectively.","description":"SmartWed 360 is a web-based wedding management marketplace that brings all essential wedding services together on one platform. It helps customers easily plan their entire wedding by browsing and booking services like wedding venues, decor, catering, photography, transportation, and accommodation, all from a single website.\r\nThe system connects customers and vendors in one place. Vendors can register and list their services by providing details such as images, pricing, and availability. The admin verifies each vendor before approving them to make sure only trusted and genuine businesses are visible on the platform.\r\nCustomers can register, log in, and explore all available services. They can filter results by location, budget, and preferences to find the most suitable options. The platform\u2019s AI Recommendation Module helps users by automatically suggesting vendors and services that best fit their requirements, saving time and effort in planning.\r\nOnce customers select their desired services, they can make bookings directly through the system. Vendors are notified about the bookings and can confirm their availability. Customers also receive updates and notifications about their reservations, ensuring smooth communication and organization.\r\nOverall, SmartWed 360 simplifies the entire wedding planning process by combining multiple services into one digital platform. It reduces stress, saves time, and provides an easy, reliable, and smart way to manage weddings, making event organization more efficient for both customers and vendors.$$\n1. Customer Module:\r\nThis module is designed for users who want to plan their wedding using the platform. Customers can register, log in, and browse all available wedding services such as venues, decor, catering, photography, transportation, and accommodation. They can view vendor details, compare packages, check availability, and make bookings according to their needs and budget. Customers also receive notifications about booking confirmations and updates. This module ensures an easy and smooth experience for users who want to manage everything in one place.\r\n2. Vendor Module:\r\nThis module is for vendors who provide wedding-related services. They can register, create profiles, and list services with images, descriptions, pricing, and availability. Vendors can receive booking requests, accept or reject them, and update their availability. This module helps vendors grow their business by reaching more customers through a digital platform without any additional marketing.\r\n3. Admin Module:\r\nThe admin module manages and monitors the entire system. Admins verify vendor registrations to ensure only authentic and high-quality vendors are approved. They handle user data, manage complaints, and remove fake or inactive listings. Admins ensure smooth communication between customers and vendors while maintaining transparency and system performance.\r\n4. Wedding Venue Module:\r\nThis module allows vendors to list wedding halls and marquees with details such as location, capacity, price (menu-based or per-person), and images. Customers can browse, compare, and book venues based on guest count, location, and budget, making hall booking simple and convenient.\r\n5. Decor Module:\r\nThis module manages decoration services like floral designs, stage setups, lighting, and theme-based decorations. Vendors can upload decor photos with pricing and details. Customers can view and choose decoration themes according to their event preferences, helping them visualize the final setup before booking.\r\n6. Catering Module:\r\nThe catering module provides food service listings with menu details, prices, and packages (buffet or per-person). Customers can browse menus, compare prices, and choose services that fit their taste and budget, simplifying food planning.\r\n7. Photography Module:\r\nThis module connects users with professional photographers and videographers. Vendors can upload their portfolio, prices, and packages. Customers can explore portfolios, compare options, and book the photographer that suits their event style and budget.\r\n8. Transportation & Accommodation Module:\r\nThis module manages travel and accommodation for wedding guests. Vendors can list cars, vans, buses, hotels, and guesthouses with pricing and availability. Customers can book vehicles for guests or bridal entries and reserve rooms for out-of-town visitors.\r\n9. AI Recommendation Module:\r\nThis module uses Artificial Intelligence to suggest vendors based on budget, location, and guest count. It analyzes preferences and automatically recommends the best available options, saving time and improving decision-making.$$\nMubashra Tahir (CIIT\/FA22-BSE-034\/WAH) is responsible for designing and developing the frontend of SmartWed 360 using React.js and Tailwind CSS.\r\nShe will create responsive pages, connect them with backend APIs, and design a user-friendly interface for customers, vendors, and admins.\r\nModules Assigned:\r\n1. Wedding Venue Module (Frontend): Designs pages for browsing and booking wedding halls and marquees.\r\n2. Decor Module (Frontend): Develops interactive UI for selecting d\u00e9cor themes and viewing images.\r\n3. Catering Module (Frontend): Creates pages for displaying menus, food packages, and prices.\r\n4. Photography Module (Frontend): Builds portfolio and photographer booking interface.\r\n5. Transportation & Accommodation Module (Frontend): Designs pages for booking vehicles and hotel stays for guests.\r\n6. AI Recommendation Module (Frontend): Integrates AI-generated vendor suggestions into the interface for smart decision-making.$$\nMuhammad Umer (CIIT\/FA22-BSE-058\/WAH) is responsible for developing the backend of the SmartWed 360 system using Node.js and Express.js. \r\nHe will manage the MongoDB database, develop and test REST APIs, and ensure secure and efficient data flow between users, vendors, and admins.\r\nHe is also responsible for backend validation, error handling, and overall system performance.\r\nModules Assigned:\r\n1. Customer Module (Backend): Handles registration, login, and booking logic for customers.\r\n2. Vendor Module (Backend): Manages vendor registration, service listings, and booking requests.\r\n3. Admin Module (Backend): Implements admin functions like vendor approval, user management, and complaint handling.\r\n4. Database Design: Creates MongoDB schemas and collections for all system entities (customers, vendors, bookings, venues, etc.).\r\n5. Integration Module: Connects frontend (React.js) with backend (Node.js) APIs, performs end-to-end testing, and ensures all modules work together seamlessly.$$\n$$\nWedBliss \u2013 Everything You Need for Your Wedding in One Place$$\nAI-Powered Vendor Recommendation System$$\nCentralized Marketplace for Wedding Services$$\nModern Full-Stack Integration (MERN & AI)","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-08 22:02:40","updated_at":"2025-10-20 12:35:31","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1027,"project_id":1405,"title":"BeadAura Handmade - customized jewellery ecommerce website","prob":"Most online jewelry stores offer only pre-designed products, limiting customers who want personalized and creative jewelry. Customers often struggle to visualize their custom designs before purchase, leading to dissatisfaction or mismatched expectations. On the other hand, small handmade jewelry sellers lack a proper digital platform to receive and manage custom orders efficiently. This project aims to solve these problems by developing an interactive jewelry design and ordering platform where customers can create their own beaded jewelry using a visual design tool, preview it in real time, and place orders directly with artisans. The system will ensure better communication between customers and sellers, reduce design errors, and empower local artisans by giving them wider market access.","description":"This project focuses on developing an interactive web-based platform that allows customers to design, customize, and order handmade beaded jewelry directly from artisans. The platform aims to bridge the gap between jewelry lovers who seek personalized accessories and small-scale artisans who create unique handmade products but lack access to digital tools and customers.\r\nThe system will feature a visual jewelry design studio, where users can choose jewelry types such as bracelets, necklaces, or earrings, and customize bead colors, materials, patterns, and sizes. A real-time preview will help users visualize their design in either 2D or 3D before placing an order. Once a design is finalized, it will be sent to registered artisans who can review, price, and accept the order for production.\r\nThe platform will include separate modules for users, sellers, and administrators. Users can manage profiles, save designs, and track orders. Sellers can upload their designs, manage custom orders, and communicate with customers. The admin will oversee user activity, approve sellers, and manage payments and disputes.\r\nBy integrating features such as customization, visualization, and marketplace connectivity, this project provides a complete digital solution for personalized jewelry shopping. It reduces miscommunication, minimizes production errors, and supports local artisans by expanding their market reach. Ultimately, the platform promotes creativity, personalization, and empowerment in the handmade jewelry industry.$$\n1. User Module\r\nThis module manages customer registration, login, and profile details. Users can design custom jewelry, browse existing collections, place orders, and track their order status. It also allows users to manage saved designs, wishlists, and communication with sellers for order updates or clarifications.\r\n2. Seller Module\r\nThis module is built for artisans and jewelry makers. Sellers can register, create shop profiles, upload jewelry templates, and manage incoming custom orders. They can view customer designs, set pricing, update order progress, and handle delivery status efficiently.\r\n3. Jewelry Design Module\r\nThe core feature of the platform, this module provides an interactive jewelry design studio where users can create their own beaded jewelry by selecting bead shapes, colors, patterns, and materials. It allows customization of jewelry type (bracelet, necklace, etc.) and stores the design data for production.\r\n4. 3D Visualization Module\r\nThis module enhances the user experience by enabling real-time 3D previews of jewelry designs. It uses 3D rendering libraries such as Three.js or Babylon.js to visualize how the customized jewelry will look when worn. It may also include AR-based previews for realistic try-on experiences on app.\r\n5. Order Management Module\r\nThis module handles the complete process of placing, tracking, and managing orders. Both customers and sellers can view order history, status updates, and notifications. It ensures smooth communication and efficient workflow between buyers and artisans.\r\n6. Payment Module\r\nThe payment module securely manages all financial transactions between customers and sellers. It supports multiple payment methods (credit\/debit cards, digital wallets, etc.) using gateways like Stripe or PayPal. It also handles refunds, receipts, and transaction history.\r\n7. Admin Module\r\nThe admin module oversees all system operations. Administrators can verify and approve sellers, manage users, monitor transactions, handle disputes, and maintain product categories. It ensures platform security, proper functioning, and fair trade between users and sellers.\r\n8. Feedback & Rating Module\r\nThis module allows users to rate their experience and provide feedback on jewelry quality and seller service. It helps maintain platform credibility, guides new customers, and encourages sellers to improve quality and reliability. \r\n9. Profile Management Module\r\nAllows both users and sellers to manage their personal and professional details.\r\n10. Product Management Module\r\nThe Product Management Module allows sellers (artisans) to efficiently manage all the handmade jewelry products they offer on the platform.$$\nIn BeadAura Handmade I shall develop modules including (i) User Module (ii) Seller Module (iii) Jewelry Design Module (iv) 3D Visualization Module (v) Product and Catalog Management Module$$\nIn BeadAura Handmade I shall develop modules including (i) Order Management Module (ii) Admin Module (iii) Payment Module (iv)Feedback & Rating Module (v) Profile Management Module$$\n$$\nJEWELLERY SHOP MANAGEMENT SYSTEM$$\nJewelry Design Module$$\n3D Visualization Module$$\nFeedback & Rating Module","comments":" $$ BeadAura Handmade - customized jewellery ecommerce website approved without AR and VR","isDraft":0,"status":2,"created_at":"2025-10-15 21:25:36","updated_at":"2025-11-02 18:27:49","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":987,"project_id":1373,"title":"EssenHub","prob":"In Germany, food ordering faces multiple challenges including lack of centralized platforms, limited payment options, and no real-time order tracking. Customers often struggle to compare the restaurants, explore diverse menus, and receive accurate delivery updates. Restaurant owners lack proper digital management tools to handle menus, sales, and customer orders efficiently. Delivery riders face coordination issues with no proper system for assigning and tracking orders. As a result, customer satisfaction, restaurant management, and delivery operations are all negatively affected.\r\nOur proposed system addresses these problems by introducing a unified web and mobile platform. It enables customers to browse restaurants, order food, make secure payments, and track orders live. Restaurant owners can manage their profiles, update menus, and monitor sales. Delivery riders can view assigned deliveries, update statuses, and navigate efficiently with interactive maps and optimized delivery routes support. This platform ensures a seamless, transparent, and reliable food ordering and delivery experience.","description":"The proposed system is a web and mobile-based food ordering and delivery platform. It bridges the gap between customers, restaurants, and delivery riders by providing a unified platform with integrated features.\r\n\r\nFor Customers: Users can sign up, browse restaurants, view menus with images, apply filters, add items to cart, and place orders. Payments can be made via multiple gateways or Cash on Delivery. Orders can be tracked live with real-time status updates and GPS-based delivery tracking. Customers can also view their order history, give ratings, and write reviews. Personalized recommendations based on preferences and order history will further enhance the customer experience. The platform also supports discount codes, loyalty points, and promotional offers to encourage repeated usage, ensuring greater customer engagement and long-term retention.\r\n\r\nFor Restaurants: Owners can manage their profiles, upload\/update menus, process incoming orders, and track daily sales reports. They can also offer discounts, manage availability, and monitor customer feedback to improve service quality. Restaurants benefit from increased visibility, automated order handling, and a wider customer base, ultimately leading to improved profitability and operational efficiency.\r\n\r\nFor Delivery Riders: Riders can accept delivery requests, update delivery progress, and mark orders as completed. GPS navigation support ensures timely deliveries, while notifications keep them updated on new orders. The system optimizes delivery routes, reducing delays and improving efficiency while boosting rider productivity.\r\n\r\nFor Admin: A web-based admin panel enables user management, restaurant approvals, complaint handling, and system oversight. Admins can generate reports, analyze performance, and maintain security across the platform. Data analytics allow tracking of sales, revenue, and user behavior for better decision-making, ensuring the overall growth and sustainability of the system.\r\n\r\nThe platform ensures secure payments, real-time tracking, push notifications, and a smooth user experience, making it a reliable, scalable, and future-ready solution for modern food delivery and restaurant management.$$\n1. Customer Module\r\nThe Customer Module provides an intuitive and user-friendly experience for individuals ordering food through the platform. Customers can register or log in using email, phone, or social media accounts. Once authenticated, they gain access to partnered restaurants, organized by cuisine, ratings, and price. Menus display images and details, while filters such as dietary needs, categories, and delivery time help refine choices. Customers can add items to the cart, apply promo codes, and pay securely via multiple gateways or Cash on Delivery. Orders are tracked in real time with live status updates, estimated delivery times, and rider details. Past orders can be viewed, re-ordered, and rated. Loyalty rewards, personalized offers, seasonal discounts, and AI recommendations enhance customer retention and satisfaction.\r\n2. Restaurant Module\r\nThe Restaurant Module empowers owners with full digital control. After admin approval, they can edit profiles, upload branding details, and set operating hours. The menu management system supports adding, updating, or removing items with descriptions, images, and pricing. Orders appear in a dashboard with instant notifications. Restaurants can accept, reject, or mark orders as complete. Sales reports are available daily, weekly, and monthly, helping analyze customer demand. Owners can launch discounts, manage reviews, and use analytics to identify trends. By digitizing operations, restaurants reduce errors, increase efficiency, and gain access to a broader market. Features like performance insights and automated inventory updates ensure streamlined management.\r\n3. Delivery Rider Module\r\nThe Delivery Rider Module ensures timely deliveries. Riders log in to receive order requests with pickup and delivery details. They can accept or decline based on availability. Accepted orders integrate with GPS navigation for optimized routes. Riders update statuses such as \u201cPicked Up,\u201d \u201cOn the Way,\u201d or \u201cDelivered.\u201d History is logged for tracking performance and accountability. Route optimization minimizes delays and saves fuel. Automatic order assignment based on location reduces idle time and maximizes deliveries. Notifications guide riders through assignments, changes, and instructions, ensuring seamless coordination and service quality.\r\n4. Admin Panel (Web)\r\nThe Admin Panel is the platform\u2019s backbone. Admins approve restaurant registrations, manage users, and resolve complaints. Fraudulent or inactive accounts can be suspended. Payment management ensures secure transactions, refunds, and gateway integration. Reports generate insights into revenue, order volume, and activity rates. Admins monitor server health, maintain data security, and enforce compliance. Complaint-handling ensures issues are addressed quickly, maintaining platform credibility. By overseeing all stakeholders, admins maintain smooth and efficient operations. Role-based access controls and audit logs strengthen transparency and platform reliability.\r\n5. Notification & Tracking Module\r\nThis module ensures real-time communication and transparency. Customers receive notifications for order placement, preparation, dispatch, and delivery. Riders are alerted to new requests and updates. Restaurants get notifications for order confirmations, payments, and reviews. GPS integration enables live tracking of riders, showing customers exact location and estimated time. Automated reminders, promotional alerts, and payment confirmations are sent across devices. These features build trust, improve satisfaction, and create a reliable ecosystem. Continuous improvements and predictive alerts ensure notifications remain timely, relevant, and effective for all stakeholders.$$\nI will develop the Mobile Application, focusing on customer interaction and delivery features. This includes user onboarding, restaurant browsing, cart & checkout, secure payment integration, GPS-based tracking, reviews, and push notifications. I will ensure a seamless and intuitive user interface, allowing customers to easily explore restaurants, filter menus, and view detailed item descriptions with images. The checkout process will support multiple payment gateways as well as Cash on Delivery, ensuring flexibility and security. Real-time order tracking with GPS integration will allow users to monitor their deliveries accurately. Additionally, I will implement a review and rating system, enabling customers to provide feedback on the food quality and delivery service. Push notifications will inform users about order updates, promotions, and personalized offers. I will also optimize the performance, handle error management, and ensure the app is responsive across different screen sizes and mobile platforms, ultimately providing a reliable and engaging user experience.$$\nI will develop the Web Application, focusing on restaurant and admin functionalities. This includes restaurant listing, menu management, order handling, admin control panel, complaint management, and sales reporting. For restaurants, I will implement features to create and update profiles, add or modify menus with detailed descriptions, prices, and images, and manage incoming orders efficiently. Restaurants will receive real-time notifications for new orders, enabling timely acceptance or rejection based on availability. For the admin panel, I will design comprehensive dashboards to monitor all platform activities, including user management, restaurant approvals, complaint resolution, and order oversight. Detailed sales reports will be provide insights into daily, weekly, and monthly performance, helping restaurants to optimize their overall operations. Secure payment processing and refund management will also be integrated. Additionally, I will ensure role-based access controls, system security, and seamless coordination between all stakeholders, providing a fully reliable and efficient platform for food delivery operations.$$\n$$\nFood Delivering App (DE-Machoos), Smart Food Ordering & Reporting System, Desire Food, Fast Food E-Ordering System for Domino\u2019s,$$\nDisplays nutrition and calorie details to help users make informed, healthier meal choices, with support for Urdu, English, and Dutch languages.$$\nOffers real-time GPS tracking, rider management, route optimization, and push notifications for faster and more transparent deliveries.$$\nUses Al to recommend meals based on user preferences and history, includes a chatbot for order confirmation, and provides an integrated complaint management system","comments":"","isDraft":0,"status":2,"created_at":"2025-10-07 14:52:55","updated_at":"2025-10-17 08:50:01","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":988,"project_id":1420,"title":"FocusMate \r\nSmart Study-Lock with Productivity Enhancement & Parental Features","prob":"Smartphone overuse and app addiction cause distractions that hurt focus,\r\nreduce productivity, and weaken study habits, especially for students.\r\nParents also struggle to manage children\u2019s screen time effectively. Current\r\napp-lock and screen-time solutions are boring, lack engagement, and fail to\r\nconnect productivity with actual learning. FocusMate addresses this by\r\nintroducing a companion mode, which will have partial control and can\r\nremotely control the locking and unlocking of apps and view the app usage\r\nanalytics. It also includes study-based lock, reminders, usage analytics,\r\nparental control(full control of child\u2019s usage). The application not only prevents distraction but also motivates users to study, focus, and maintain digital balance, making it\r\nuseful for students, parents, and general users.","description":"FocusMate is a mobile productivity and parental control app that minimizes\r\ndistractions and promotes disciplined device use through its core feature,\r\nCompanion Mode. Users can connect with a trusted companion, such as a\r\nparent, friend, professional or mentor, who can remotely lock or unlock apps, monitor\r\nscreen-time analytics, and approve or deny unlock requests. Its\r\nParental Control module gives parents full remote authority to manage their\r\nchild\u2019s device and view detailed usage reports without physical access. The app also includes a\r\ndedicated study workspace with AI-powered summaries and quick quizzes,\r\nalong with usage analytics and reminders to visualize and improve digital\r\nhabits. With secure authentication and accountability features, FocusMate\r\nhelps students stay focused, parents stay in control, and users maintain a\r\nhealthy, balanced relationship with technology.$$\n1. Authentication Module\r\nThis module manages secure login, signup, and logout for users, parent and companion. It prevents bypassing app locks and ensures that only authorized users can access features.\r\nBasic password recovery and session management will also be supported to\r\nmaintain security.\r\n\r\n2. Companion Control mode\r\nThis module ensure that you can give access to\r\nany companion i.e. your friend, mentor, teacher or a professional etc. that will remotely monitor and control your mobile app usage and locks. \r\n\r\n3. Study-Pass Engine\r\nThis is the module that directly connects studying with digital access. Users\r\ncan upload study material select difficulty levels. The engine will generate study tasks and auto-grade the responses. Only when the user attempts the quiz will the restricted apps be unlocked.\r\n\r\n4. Study Workspace\r\nThe Study Workspace provides an interactive learning environment inside the app.\r\nIt includes embedded viewers for PDFs. This space ensures that\r\nstudents don\u2019t need to switch to external apps, keeping them focused and\r\nengaged.\r\n\r\n5. App Lock\r\nThis module enforces app restrictions by two methods based on access:\r\nCompanion\/Parent's Access: Companion\/Parent controls which apps to lock.\r\nNormal mode: Allows manual and scheduled locks with user control.\r\n\r\n6. Parental Control Module\r\nParents can apply settings such as restricted or unrestricted apps remotely and can also monitor user activities. After setting the lock, it will only be unlocked by parents.\r\n\r\n7. Analytics & Reminders Module\r\nThis module tracks usage patterns, generates reports, and visualizes statistics such\r\nas screen time, study duration, and app usage frequency. For students, it\r\nhighlights study consistency; for parents, it shows clear progress and overuse\r\npatterns. Notifications are sent to reinforce responsible device use.$$\nApp lock system (time-based and manual app locks) \r\nParental control (configure study sessions, set locks on same device) \r\nReminders and analytics (break alerts, overuse warnings, usage reports, graphical \r\nstatistics) \r\nAssessment engine (Auto-grading and unlock criteria)$$\nStudy-Pass Engine (upload study material, topic & difficulty selection summaries, \r\nquiz\/assignment creation) \r\nEmbedded study workspace (PDF\/image viewer, interactive notes) \r\nCompanion mode (Partial control of user activities)$$\n$$\n$$\n$$\n$$\n","comments":" $$ Approved with updated scope","isDraft":0,"status":2,"created_at":"2025-10-07 15:24:36","updated_at":"2025-10-20 12:45:50","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1038,"project_id":1321,"title":"VeriFy \u2013 Multi-Source AI-Powered Fake News Detection and Verification System with Insights and Cross-Domain Analysis","prob":"In today\u2019s digital age, misinformation spreads faster than truth, especially across social media platforms and unverified news websites. People often fall for fake or manipulated news that influences opinions, politics, health decisions, and public trust. Existing fake news detection systems usually analyze single articles without verifying them across multiple sources or explaining why they were classified as fake. Our project, VeriFy, addresses this by using Artificial Intelligence, Natural Language Processing (NLP), and Explainable AI (XAI) to detect, compare, and justify news authenticity. It cross-checks news across multiple reliable sources like BBC, Reuters, and Dawn using APIs, evaluates how news spreads on platforms like Twitter and Reddit, and identifies credibility differences across domains such as politics, health, and entertainment. VeriFy aims to combat misinformation by providing transparent, data-driven insights that help users, journalists, and fact-checkers easily identify false or misleading content in real time.","description":"VeriFy is an AI-powered web-based system designed to verify the authenticity of online news articles through cross-source analysis, NLP pipelines, and Explainable AI. Users can paste a news URL or article text into VeriFy, which then fetches related news reports from trusted sources using APIs such as Google News API and NewsAPI. It compares the similarity of headlines, content, and sentiment across multiple outlets like BBC, Reuters, and Dawn. The system also integrates data from Twitter (X) and Reddit to analyze public discussion trends and sentiment related to the same topic.\r\n\r\nUsing multilingual NLP models (English and Urdu), VeriFy preprocesses text and applies a hybrid fake news classification model built with BERT, LSTM, and Logistic Regression to assess credibility. The system explains its results using Explainable AI techniques, displaying phrases or patterns that led to a \u201cFake\u201d or \u201cReal\u201d prediction.\r\n\r\nUsers can view results on a dashboard or through a Chrome Extension, which automatically highlights fake news probability when browsing articles. VeriFy also includes a domain credibility database that ranks news categories based on historical accuracy. Additionally, users can receive notifications or emails when verified fake news articles match ones they\u2019ve read.\r\n\r\nThe system continuously learns from user feedback and updates its models using new verified datasets. VeriFy combines transparency, multilingual support, real-time fact-checking, and explainable reasoning\u2014providing an intelligent and trustworthy tool to help society fight misinformation effectively.$$\n1. Login & Signup Module:\r\nAllows secure user registration and authentication using encrypted credentials. Users can save verification history and manage preferences for alerts and notifications.\r\n\r\n2. News Input Module:\r\nLets users enter a URL or paste article text. Automatically detects article language (English or Urdu) and initiates the verification process.\r\n\r\n3. Preprocessing & NLP Module:\r\nCleans and tokenizes text, removes noise, and performs stemming\/lemmatization. Supports multilingual text processing using language-specific pipelines (English and Urdu).\r\n\r\n4. Fake News Classification Module:\r\nUses hybrid machine learning (BERT, LSTM, Logistic Regression) for determining if the news is real or fake. Outputs prediction along with confidence percentage.\r\n\r\n5. Explainable AI (XAI) Module:\r\nHighlights words or phrases that influenced the classification decision, ensuring transparency and interpretability in predictions.\r\n\r\n6. Cross-Verification Module:\r\nFetches similar news from APIs (Google News, NewsAPI) and compares semantic similarity using Sentence-BERT. Detects inconsistencies across reliable outlets.\r\n\r\n7. Multi-Source Article Fetching Module:\r\nPulls related news from multiple platforms and regions to determine if similar content is reported by verified media.\r\n\r\n8. Social Media Trend Analysis Module:\r\nAnalyzes how the same topic spreads across Twitter\/X and Reddit. Evaluates sentiment and frequency to detect potential misinformation trends.\r\n\r\n9. Domain Credibility Module:\r\nMaintains a domain-wise credibility score (Politics, Health, Entertainment, etc.) based on historical accuracy and fake news frequency.\r\n\r\n10. Results Visualization Module:\r\nDisplays results using charts and visual metrics: (i) Real\/Fake label, (ii) Confidence score, (iii) Evidence phrases, (iv) Related verified sources.\r\n\r\n11. Chrome Extension \/ Dashboard Module:\r\nAllows users to check news authenticity directly from their browser. Shows a pop-up with probability score and link to detailed analysis.\r\n\r\n12. Alert & Notification Module:\r\nSends real-time alerts via email or push notifications if news verified as fake matches previously viewed content.\r\n\r\n13. Feedback Module:\r\nAllows users to provide feedback on predictions (correct or incorrect). Collected data helps refine model accuracy.\r\n\r\n14. Admin Dashboard Module:\r\nEnables administrators to monitor API calls, user activity, model accuracy, and retraining processes.\r\n\r\n15. Retraining & Update Module:\r\nAutomatically updates the fake news detection model using new verified datasets and user feedback to maintain accuracy.$$\nI will develop the Login & Signup, News Input, Cross-Verification, Multi-Source Fetching, Chrome Extension\/Dashboard, Alert & Notification, Feedback, and Admin Dashboard modules, focusing on frontend, backend, and user experience components.$$\nI will develop the Preprocessing & NLP, Fake News Classification, Explainable AI (XAI), Results Visualization, Domain Credibility, Retraining & Update, and Social Media Trend Analysis modules, focusing on the AI pipelines, model development, and explainable reasoning mechanisms.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-21 20:37:50","updated_at":"2025-11-04 12:54:05","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":995,"project_id":1342,"title":"SnapLaw: One Tap to Legal Clarity","prob":"The project addresses key challenges in the legal system, such as limited access to justice, complex procedures, and lack of public legal awareness. Clients struggle to verify contracts and find reliable lawyers, while lawyers face inefficiencies in research and case management. Citizens often remain unaware of their rights, leading to confusion and vulnerability.\r\nThis FYP introduces an AI-powered mobile application connecting clients, lawyers, and citizens on one platform. It offers features like Contract Risk Radar, LawBot Q&A, Precedent Finder, Case Tracker, and Citizen Guidance, along with SafeSpeak (Abuse Guidance), Trusted Booking, Secure Evidence Scanner, and Privacy Vault for safe and transparent interactions.\r\nUnlike general AI tools like ChatGPT, which provide generic responses, our system uses LegalBERT and RoBERTa, fine-tuned on Pakistani legal data, to deliver law-specific, context-aware, and accurate insights. It focuses on legal interpretation and compliance, ensuring precision, reliability, and ethical alignment for real legal applications.","description":"This FYP focuses on developing an AI-powered mobile application that connects clients, lawyers, and citizens within a unified digital ecosystem to improve access to legal resources and services. The system leverages advanced Natural Language Processing (NLP) models and intelligent tools to simplify legal understanding, automate assistance, and enhance transparency.\r\nClients can post cases, upload documents, and analyze contracts through the Contract Risk Radar for clause detection. They can also track case progress, book appointments via the Trusted Booking & Calendar, scan evidence securely using OCR-based redaction, and store files safely in the Privacy Vault.\r\nLawyers manage cases through the Case Dashboard, perform AI-based precedent searches, summarize judgments, and compare case notes for better preparation. They also receive ratings and feedback from clients to promote trust and transparency.\r\nCitizens gain awareness through Guidance Modules, Gamified Legal Learning, and a Citizen Justice Tracker showing real-time case duration statistics.\r\nAt the core lies the intelligent LawBot, offering multilingual legal Q&A, SafeSpeak (Abuse Guidance), Bias Detection, and Contract Simplification using models like LegalBERT and RoBERTa.\r\nOverall, the system empowers citizens with legal literacy, assists clients with secure AI-driven insights, and supports lawyers with advanced research tools\u2014creating a fair, accessible, and tech-driven legal ecosystem.$$\n1. Client Modules: Case posting, document upload, Contract Risk Radar, Case Progress Tracker, Trusted Booking & Calendar, Secure Evidence Scanner, Privacy Vault.\r\n2. Lawyer Modules: Case Dashboard, Legal Precedent Finder, Case Summarizer, Study Journal & Case Comparison, Ratings & Reviews.\r\n3. Citizen Modules: Citizen Guidance Laws, Citizen Justice Tracker, Gamified Legal Awareness, Offline Mode (Lite).\r\n4. LawBot Modules: Legal Q&A, SafeSpeak (Abuse Guidance), Bias Checker, Contract Simplifier, Multilingual Support.\r\n5. Admin Module: Lawyer Verification, Abuse Monitoring, Citizen Law Library Updates, Dataset Management.$$\n(i) Contract Risk Radar \u2013 Analyzes uploaded contracts to detect unfair or risky clauses using AI (LegalBERT).\r\n(ii) Case Summarizer \u2013 Uses NLP models to summarize long judgments or legal documents into easy language.\r\n(iii) Trusted Booking & Calendar \u2013 Allows one-tap lawyer appointments (in-person\/video) with auto court-date alerts.\r\n(iv) Secure Evidence Scanner \u2013 Scans FIRs, bills, or photos using OCR, redacting sensitive data before sharing.\r\n(v) Privacy Vault \u2013 Provides end-to-end encrypted storage for sensitive contracts and legal PDFs.$$\n(i) Lawyer Dashboard \u2013 Manages client cases, hearing updates, and personal schedules.\r\n(ii) Legal Precedent Finder \u2013 Uses AI text search to find relevant judgments and references.\r\n(iii) LawBot \u2013 Chatbot using NLP for real-time legal Q&A, advice, and contract simplification.\r\n(iv) Citizen Guidance & Justice Tracker \u2013 Displays legal awareness guides and case resolution stats.\r\n(v) SafeSpeak (Abuse Guidance) \u2013 Users report abuse or harassment; LawBot provides legal guidance and reporting steps.\r\n(vi )Offline Mode (Lite) \u2013 Allows users to access saved chats and guides without an internet connection.\r\n(vii) Ratings & Reviews (Post-Case) \u2013 Enables feedback on lawyer performance, with moderation for authenticity.$$\n$$\nCiviLex: Judiciary Management System with AI-Powered Drafting and Land Revenue Integration$$\nTheir project automates court and judiciary operations, while ours focuses on AI-driven legal assistance and public awareness for citizens, clients, and lawyers$$\nThey use Google Gemini for document drafting; we use LegalBERT and RoBERTa, fine-tuned on Pakistani legal data, for intelligent understanding, risk detection, and legal Q&A.$$\nTheir system is web-based for judges and admins; ours is a mobile app empowering the public with tools like LawBot, Contract Risk Radar, SafeSpeak, and Citizen Guidance.","comments":"","isDraft":0,"status":2,"created_at":"2025-10-07 23:11:56","updated_at":"2025-11-05 09:09:43","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}
No. Project Title Reg. No. Name Date & Time Evaluation Status
1. Echoes of Isolation – A Voice-Responsive Paranormal Horror Game CIIT/FA22-BCS-006/WAH
CIIT/FA22-BCS-032/WAH
 MUHAMMAD HUZAIFA MUMTAZ
ABDULLAH KHALIL
 Done Accepted  
2. NextPakistan – Empowering AI Talent, Solving Real-World Problems CIIT/FA22-BSE-020/WAH
CIIT/FA22-BSE-049/WAH
 MUHAMMAD AHMAD RANA
WAQEEQ BIN HASSAN
 Done Accepted  
3. A Web-Based Digitalization Toolkit for Startup Workflow Automation CIIT/FA22-BCS-113/WAH
CIIT/FA22-BCS-125/WAH
 SARSHAAR FATIMA
HAFSA NAZIR
 Done Accepted  
4. MediusCare: An AI-Powered Health Monitoring and Assistance System for Diabetes and Hypertension Patients CIIT/FA22-BSE-065/WAH
CIIT/FA22-BSE-066/WAH
 ALI HASSAN
MUHAMMAD ARSLAN
 Done Accepted  
5. AniAd – Turn your script into animation CIIT/FA22-BSE-024/WAH
CIIT/FA22-BSE-085/WAH
 MUHAMMAD HUSNAIN SHEHNSHA
FAKHIR HASSAN SHABA
 Done Accepted  
6. Intelligent Web Application For Brand Collaboration And Software project Management CIIT/FA22-BCS-046/WAH
CIIT/FA22-BCS-087/WAH
 MUHAMMAD BIN MEHMOOD
MUHAMMAD SHAHZAIB
 Done Accepted  
7. AI-Powered Real-Time Telephony Bridge with Analytics for Call Automation CIIT/FA22-BCS-023/WAH
CIIT/FA22-BCS-030/WAH
 MUHAMMAD MAHDI ALI MIR
MUHAMMAD HARIS JAVED
 Done Accepted  
8. Sales Distribution Hub CIIT/FA22-BSE-045/WAH
CIIT/FA22-BSE-122/WAH
 BILAL ARSHAD
MUZZAIB SAJJAD
 Done Accepted  
9. CiviLex: Judiciary Management System with AI-Powered Drafting and Land Revenue Integration CIIT/FA22-BSE-094/WAH
CIIT/FA22-BSE-119/WAH
 SIDRA MALIK
SHAHZEB SHOUKAT
 Done Accepted  
10. Tale-Toon: Turn Tales into Toons for Kids. CIIT/FA22-BCS-048/WAH
CIIT/FA22-BCS-052/WAH
 TALHA NAVEED
ABDULLAH NADEEM
 Done Accepted  
11. AI-Powered Mannequin-Based Medical Training System (AIMMS) CIIT/FA22-BCS-019/WAH
CIIT/FA22-BCS-026/WAH
 ZAHRA NAVEED
ZAINAB SIDDIQUI
 Done Accepted  
12. Lodge Logic: Comprehensive Guest House Management System CIIT/FA22-BCS-060/WAH
CIIT/FA22-BCS-089/WAH
 HASSAAN HABIB
MOHAMMAD ALI AWAN
 Done Accepted  
13. RoomLoom - An AI-Powered Furniture Advisor CIIT/FA22-BSE-069/WAH
CIIT/FA22-BSE-121/WAH
 AREEBA TARIQ
SAMMAN HABIB
 Done Accepted  
14. Brain-Computer Interface for Paralyzed Patients: Real-Time Affective States Monitoring using NeuroSky CIIT/FA22-BCS-016/WAH
CIIT/FA22-BCS-017/WAH
 SUMRA ZAMAN
SALIHA BIBI
 Done Accepted  
15. UniLearn AI—university + learning powered by AI. CIIT/FA22-BSE-108/WAH
CIIT/FA22-BSE-115/WAH
 MUHAMMAD AMMAR
MUHAMMAD MAAZ SHAH
 Done Accepted  
16. VissaAssist CRM CIIT/FA22-BCS-002/WAH
CIIT/FA22-BSE-072/WAH
 LARAIB
MAHAM ASHRAF
 Done Accepted  
17. DeceptiVerse: AI-Powered Dynamic Deception Framework Using Fake Web Applications and Deepfake Audio CIIT/FA22-BCS-066/WAH
CIIT/FA22-BCS-067/WAH
 MUSTAJAB ULLAH KHAN
UMER ZAHEER LODHI
 Done Accepted  
18. AI-Powered Clothing Brand with Voice Assistant and Smart Inventory Management CIIT/FA22-BCS-062/WAH
CIIT/FA22-BCS-156/WAH
 KHUBAIB YOUSUF
MUHAMMAD MESUM
 Done Accepted  
19. Groomio: Where Style Begins – A Smart Barber Booking & AI Powered Grooming Recommendation Platform CIIT/FA22-BSE-011/WAH
CIIT/FA22-BSE-018/WAH
 MUHAMMAD MUNEEB-UR-REHMAN
SAIF UL HASSAN
 Done Accepted  
20. Ustad Go: AI-powered platform for hiring and managing verified local service providers. CIIT/FA22-BCS-061/WAH
CIIT/FA22-BCS-170/WAH
 MUHAMMAD UMER MEHBOOB
MUHAMMAD AHMAD MUDASSAR
 Done Accepted  
21. AI-Powered Adaptive Learning Platform with Personalized Pathways CIIT/FA22-BCS-028/WAH
CIIT/FA22-BCS-111/WAH
 ZAIN UL ABIDEEN
MUHAMMAD AHMAD TARIQ
 Done Accepted  
22. EduBite A Smart Digital Marketplace for School/College & University Canteens CIIT/FA22-BCS-101/WAH
CIIT/FA22-BSE-029/WAH
 IBRAR AHMAD
NAILA BIBI
 Done Accepted  
23. DupeFinder: Affordable Alternatives for Luxury Wearables CIIT/FA22-BCS-088/WAH
CIIT/FA22-BCS-132/WAH
 AROOJ KHAN
MUHAMMAD USAMA INAYAT
 Done Accepted  
24. AgriVision: AI Diagnosis and Orchard Trade Hub CIIT/FA22-BCS-118/WAH
CIIT/FA22-BCS-139/WAH
 AQIB SHEHZAD
KASHIF ALI
 Done Accepted  
25. PakPulse AI: Multi-Disease Outbreak Forecasting & Early Warning System for Pakistan CIIT/FA22-BCS-044/WAH
CIIT/FA22-BCS-149/WAH
 HAMZA ANJUM KIANI
MUHAMMAD ABDULLAH SHEIKH
 Done Accepted  
26. CleanCore – AI-Driven Waste Monitoring and Optimization CIIT/FA22-BCS-009/WAH
CIIT/FA22-BCS-020/WAH
 AYESHA NOMAN
AYESHA NADEEM
 Done Accepted  
27. IoT-Enabled Smart Energy Meter with Intelligent Load Control and Utility Integration. CIIT/FA22-BCS-033/WAH
CIIT/FA22-BCS-049/WAH
 AMMARA IQBAL
ESHA TUR RAZIA
 Done Accepted  
28. ClarityWorks - Calibrated Reading Level & Smart Simplify CIIT/FA22-BSE-067/WAH
CIIT/FA22-BSE-100/WAH
 JUNAID AHSAN MALIK
MUHAMMAD WAHAJ NAVEED
 Done Accepted  
29. SchoolMate – Smart School Van Tracking & Management System CIIT/FA22-BSE-105/WAH
CIIT/SP22-BSE-053/WAH
 MUHAMMAD SAQIB
UZAIR ARIF
 Done Accepted  
30. SmartWed 360: A Marketplace for Complete Wedding Management. CIIT/FA22-BSE-034/WAH
CIIT/FA22-BSE-058/WAH
 MUBASHRA TAHIR
MUHAMMAD UMER
 Done Accepted  
31. BeadAura Handmade - customized jewellery ecommerce website CIIT/FA22-BSE-012/WAH
CIIT/FA22-BSE-041/WAH
 EZZA FATIMA
FATIMA RASHID
 Done Accepted  
32. EssenHub CIIT/FA22-BCS-133/WAH
CIIT/FA22-BCS-153/WAH
 KHADIJA SHAHID
AFIA JAHANGIR
 Done Accepted  
33. FocusMate Smart Study-Lock with Productivity Enhancement & Parental Features CIIT/FA22-BSE-037/WAH
CIIT/FA22-BSE-109/WAH
 MUAZ AHMED
HASHIR AHMAD
 Done Accepted  
34. A Smart CMS with QR-Based Attendance and Integrated Student–Teacher System CIIT/FA22-BCS-005/WAH
CIIT/FA22-BCS-036/WAH
 FAHAD SULTAN
MUHAMMAD MUNIR
 Done Accepted  
35. VeriFy – Multi-Source AI-Powered Fake News Detection and Verification System with Insights and Cross-Domain Analysis CIIT/FA22-BCS-084/WAH
CIIT/FA22-BCS-099/WAH
 RAJA MUHAMMAD UMAIR
MUHAMMAD HARIS
 Done Accepted  
36. SnapLaw: One Tap to Legal Clarity CIIT/FA22-BSE-092/WAH
CIIT/FA22-BSE-096/WAH
 QURAT-UL-AIN
ZOHA GULL
 Done Accepted  
Committee: 2
Team Members:
Dr. Tassawar Iqbal (HEAD)
Dr. Samia Riaz
Mr. Muhammad Nadeem
Mr. Hassan Sardar
Venue: CS Conference Room No. 1
Remarks: Dear Students, You all are directed to follow the plan, reach the venue on time, and avoid to queue up in the faculty hall.

1{"id":1043,"project_id":1401,"title":"AI-Based Elderly Healthcare Assistant","prob":"This project, titled AI Elderly Healthcare Assistant, addresses challenges faced by elderly people, especially those living alone. Many struggle with health management, emotional well-being, and lack of timely medical support. Current systems do not provide continuous monitoring or intelligent communication for proper mental and physical care at home.\r\n\r\nOur AI-based system serves as a virtual health companion that interacts through natural text and voice. It offers emotional support by detecting sadness or stress and keeps a personalized record of user behavior. The system includes an AI Health Risk Prediction module trained on public datasets to detect diseases such as heart problems or diabetes, generating explainable risk reports for doctors. Additionally, a webcam-based emotion and activity monitor tracks facial expressions and inactivity to identify distress or fatigue.\r\n\r\nThrough intelligent AI interaction, predictive analysis, and doctor connectivity, it ensures continuous digital care, early risk detection, and emotional companionship for elderly individuals.","description":"This healthcare system is built around three main panels: Admin Panel, Doctor Panel, and Patient Panel, each designed to ensure smooth and efficient functioning. The Admin Panel manages the entire platform, the Doctor Panel allows doctors to monitor and guide patients, and the Patient Panel enables elderly users to access health data and communicate easily with doctors.\r\n\r\nThe Admin Panel serves as the control center of the system. It manages authentication, login, logout, and role-based access for all users. The admin can add, edit, or delete doctors and patients, assign patients to specific doctors, and monitor system activity. It also tracks AI model performance, alerts, and communication logs. The admin oversees AI modules that can be trained, updated, and evaluated for accuracy. For proper oversight, the admin receives weekly reports summarizing doctor activity, patient updates, and system statistics.\r\n\r\nThe Doctor Panel helps doctors manage patients through an AI-assisted dashboard. Instead of wearable devices, doctors view AI-generated health predictions, risk scores, and emotion analysis derived from patient interactions. Doctors can review medical history, add notes, verify AI alerts, and receive instant notifications in case of critical findings. A communication module enables direct interaction via chat, call, or prescription sharing, while access to reports supports better decision-making.\r\n\r\nThe Patient Panel is simple and user-friendly. Patients interact with the AI Virtual Health Companion through text or voice to discuss health and emotions. The system offers emotional support, detects stress, and provides AI-generated health insights. In emergencies, patients can send instant alerts or use the SOS option to contact doctors. Communication features support chatting and video consultations, while the history section maintains previous reports and advice.\r\nThrough these integrated panels and advanced AI capabilities, the AI Elderly Healthcare Assistant ensures digital monitoring, emotional companionship, and timely medical assistance$$\n1. AI Virtual Health Companion\r\n\r\nThe AI Virtual Health Companion serves as the main interaction module between elderly users and the system. It replaces the need for any wearable band or physical sensors by allowing natural and interactive communication through text and voice. This intelligent chatbot uses Natural Language Processing (NLP) and Speech Recognition to understand user messages and respond in a friendly, human-like manner. The AI companion provides emotional support, detects stress or sadness from tone and conversation patterns, and remembers user preferences and routines. It encourages healthy habits, sends positive messages, and ensures that elderly users feel heard and supported. Whenever a user reports discomfort or unusual symptoms, the companion automatically flags the issue for medical review in the doctor panel, ensuring timely attention without any manual input.\r\n\r\n2. Patient Panel\r\n\r\nThe Patient Panel is designed to be simple, clear, and elderly-friendly. Instead of displaying data from wearable devices, it shows AI-generated health predictions, emotional insights, and risk levels based on patient interactions and stored information. Patients can communicate with the AI Virtual Health Companion to discuss their health, share symptoms, or express emotions. The system analyzes their behavior and identifies patterns that might indicate health issues or emotional stress.\r\n\r\nThe panel also provides intelligent notifications, such as emotional health tips, AI-based health suggestions, appointment reminders, and doctor updates. A dedicated history and reports section allows patients to review previous AI alerts, medical consultations, and track their personal progress over time.\r\n\r\n3. Doctor Panel\r\n\r\nThe Doctor Panel provides healthcare professionals with an AI-assisted dashboard for smarter monitoring and decision-making. Instead of relying on physical sensor readings, doctors access AI Health Risk Predictions, emotional analysis, and activity summaries automatically generated from patient interactions. The dashboard highlights patients with high-risk scores or signs of distress, allowing doctors to prioritize cases effectively. Doctors can verify AI alerts, review medical history, add notes, and receive AI-suggested actions during critical situations. The system also enables smooth communication with patients through chat, voice, or video calls, and supports digital prescription sharing. Doctors can download reports and track changes in patient well-being over time, enhancing both accuracy and patient care.\r\n\r\n4. Admin Panel\r\n\r\nThe Admin Panel acts as the main control unit of the platform. It manages user authentication, access roles, and data integrity for Admin, Doctor, and Patient accounts. The admin monitors overall system performance, including AI model accuracy, alert logs, and communication statistics. They can train, test, and update AI modules to improve prediction reliability and emotion detection performance. The admin also handles user management\u2014adding, editing, or deleting profiles\u2014and assigns patients to specific doctors. Regular activity reports summarizing AI model performance, doctor activity, and patient alerts are automatically generated to maintain transparency and accountability.$$\nI shall develop the modules which include:\r\n1. AI Virtual Health Companion Module\r\n\r\nThis module serves as the main interaction layer between the elderly user and the system. It enables natural communication through text or voice using AI and NLP technologies. The companion offers emotional support, detects stress or sadness from conversations, and provides personalized health suggestions. It remembers user preferences and routines, ensuring consistent engagement. If abnormal symptoms or emotional distress are detected, it notifies the doctor panel for further review.\r\n\r\n2. Patient Panel Module\r\n\r\nThe Patient Panel is simple and user-friendly, allowing secure login and interaction with the AI Virtual Health Companion. Patients receive AI-generated health insights, emotional feedback, and risk predictions based on their behavior and stored records. It delivers intelligent notifications, including AI-based tips, reminders, and doctor updates. The History and Reports section lets users review previous alerts, consultations, and progress over time.$$\nI shall develop the modules which include:\r\n3. Doctor Panel Module\r\n\r\nThe Doctor Panel provides healthcare professionals with an AI-assisted dashboard for smart monitoring and decision-making. Doctors can view AI health risk predictions, emotional analysis, and patient activity summaries. They can review medical history, verify AI alerts, and add notes. In critical cases, the system automatically sends AI-suggested actions and patient summaries. Doctors can communicate directly with patients via chat, calls, or prescription sharing, ensuring better and faster medical guidance.\r\n\r\n4. Admin Panel Module\r\n\r\nThe Admin Panel functions as the main control unit of the system. It manages authentication, user access, and role-based permissions for all accounts. The admin oversees AI module training, performance evaluation, and alert monitoring. User management tasks include adding, editing, and assigning patients to doctors. The panel controls doctor activity and patient updates for better transparency and management.$$\n$$\n$$\n$$\n$$\n","comments":" $$ The revised proposal effectively resolves earlier feasibility concerns by replacing wearable-based monitoring with an AI-driven virtual system that uses emotion and text interaction for elderly care. The modules are now clearly structured and appropriately assigned, reflecting a sound understanding of AI, NLP, and emotion detection integration. Overall, the project presents a more practical and well-defined approach, though it is recommended to include a caretaker or emergency contact notification feature for real-time alerts during critical health or distress situations.","isDraft":0,"status":2,"created_at":"2025-10-24 15:31:11","updated_at":"2025-11-08 11:50:37","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":961,"project_id":1368,"title":"Cloth Warehouse Inventory Management System","prob":"The cloth industry often struggles with managing large warehouse inventories due to\r\nmanual record-keeping, mismanagement of stock levels, and inefficient order handling.\r\nWarehouse managers face challenges like overstocking, stockouts, slow order\r\nprocessing, and difficulty in tracking purchases and sales.\r\nThe proposed system solves these problems by providing an automated digital solution\r\nthat handles inventory, product catalogs, purchase orders, and sales orders in one\r\ncentralized platform.","description":"This project is designed to create a Warehouse Inventory and Management System\r\nthat streamlines business operations by managing products, orders, stock, and\r\nrecommendations efficiently. Cloth Warehouse Inventory Management System\r\n(CWIMS) is a web-based application designed to streamline warehouse operations.\r\nThe system is structured around six key modules: 1. Inventory management module\r\nIt manages inventory updates, purchase and sales orders. 2. Product Catalog\r\nModule: This module maintains a detailed catalog of all products. It allows users to\r\neasily search, update, and organize product information, making inventory\r\nmanagement and sales more efficient. 3. Purchase Order Management Module:\r\nThis module handles all supplier orders. It enables the creation, tracking, and\r\napproval of purchase orders. 4. Sales Order Management Module: This module\r\nmanages customer orders. It tracks order status, updates stock levels accordingly,\r\nand ensures timely fulfillment to enhance customer satisfaction. 5. Return &\r\nDamage Handling Module: This module deals with product returns, damaged goods.\r\nIt updates inventory and generates reports to minimize losses and improve quality\r\ncontrol. 6. AI Recommendation and AI-Driven Demand Forecasting Module: Uses the AI Recommendation to suggest additional products at checkout. 7. Admin Control & Reporting Module: Provides a centralized management system where admin can add, update, or remove staff accounts. Overview of sales, purchases, stock levels, returns, and online orders. 8. E-Commerce & Online Storefront module extends the warehouse inventory system into a fully functional online sales platform.$$\nIn the cloth warehouse inventory management system, the module are follows: 1.\r\nInventory management Module: This module is responsible for tracking, updating,\r\nand controlling stock in the warehouse. It ensures that all clothes (fabrics, shirts,\r\ntrousers, etc.) are accurately recorded in real time, reducing human error and\r\nimproving efficiency. 2. Product Catalog Module: this module manages all product\r\ndetails like fabric type, size, color, price, and stock code. Provides quick search and\r\nfilter options (by fabric type, category, or price).\r\n3. The Purchase Order Management Module is designed to handle all interactions\r\nwith suppliers. It helps warehouse managers create, track, and manage purchase\r\norders efficiently. This module ensures that inventory is replenished on time,\r\nsuppliers are managed effectively, and stock levels are automatically updated once\r\ngoods are received. 4. Sales Order Management Module: The Sales Order\r\nManagement Module manages all customer sales transactions in the warehouse\r\nsystem. It ensures that sales orders are properly recorded, stock levels are updated\r\nin real time. 5. Return & Damage Handling Module: The Return & Damage\r\nHandling Module manages all cases where products are returned by customers or\r\nfound to be damaged\/defective in the warehouse. 6. The AI Recommendation\r\nSystem and AI driven demand forecasting is designed to suggest products to customers or ware house staff. Example: If a customer buys \u201cCotton Shirt,\u201d system suggests \u201cCotton Trouser\u201d or \u201cSummer Fabric to customer. Using historical sales data, seasonal trends, and purchase behavior, the system predict future demand for each product. It helps the warehouse avoid overstocking and understocking by suggesting optimal reorder quantities. These suggestion are based on:\r\n\u2022 Customer purchase history.\r\n\u2022 Popular\/trending items.\r\n\u2022 Related fabric types or categories. Product sales per day\/week\/month\r\n\u2022 Price at which it was sold\r\n\u2022 Time of sale (season, holidays, weekends vs weekdays)\r\nThis module helps in upselling (selling a higher range) and cross-selling (selling\r\nrelated items), improving overall sales and customer satisfaction. Provides\r\npersonal recommendations to customers and warehouse staff. . 7. Admin Control & Reporting Module: Provides a centralized management system where the admin oversees all warehouse and e-commerce operations. Admin can add, update, or remove staff accounts. Overview of sales, purchases, stock levels, returns, and online orders. 8. E-Commerce & Online Storefront module extends the warehouse inventory system into a fully functional online sales platform. It allows customers to browse products in real time, add them to a shopping cart, and securely purchase them online.$$\nIn the cloth warehouse inventory management system Project, I\r\ndevelop following modules: 1. Inventory management Module: This\r\nmodule is responsible for tracking, updating, and controlling stock in\r\nwarehouse. It ensures that all clothes are\r\naccurately recorded reducing human error and improving\r\nefficiency. 2. Product Catalog Module: this module manages all product\r\ndetails like fabric type, size, color, price, and stock code. Provides quick\r\nsearch and filter options\r\n3. The Purchase Order Management Module is designed to handle all\r\ninteractions with suppliers. It helps warehouse managers create, track, and\r\nmanage purchase orders efficiently. This module ensures that inventory is\r\nreplenished on time, suppliers are managed effectively, and stock levels are\r\nautomatically updated once goods are received. 4. E-Commerce & Online Storefront module extends the warehouse inventory system into online sales platform. It allows customers to browse products in real time, add them to a shopping cart, and securely purchase them online.$$\nIn the cloth warehouse Inventory management system Project, I shall develop the\r\nfollowing modules: 1. Sales Order Management Module: The Sales Order\r\nManagement Module manages all customer sales transactions in the warehouse\r\nsystem. It ensures that sales orders are properly recorded, stock levels are updated\r\nin real time. 2. Return & Damage Handling Module: The Return & Damage\r\nHandling Module manages all cases where products are returned by customers or\r\nfound to be damaged\/defective in the warehouse. 3. The AI Recommendation\r\nSystem and AI driven demand forecasting is designed to suggest products to customers or warehouse staff. . 4.Admin Control & Reporting Module: Provides a centralized management system where the admin oversees all warehouse and e-commerce operations. Admin can add, update, or remove staff accounts. Overview of sales, purchases, stock levels, returns, and online orders.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Students have added Super Admin support and E Commerce to the proposal as recommneded by the comitee. The proposed modules must be developed with true application of the modules","isDraft":0,"status":2,"created_at":"2025-09-29 21:45:46","updated_at":"2025-10-15 15:27:18","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":970,"project_id":1327,"title":"Insight AI: \u201cAI-Powered Data Analytics for Smarter Decisions.\u201d","prob":"In today\u2019s data-driven world, non-technical users such as managers, educators, and healthcare workers even though the technical users also struggle to analyze datasets because existing tools like SQL, Tableau, and Power BI require technical expertise and complex setups. This creates barriers to extracting timely insights, limits accessibility, and increases dependence on data analysts. Our FYP solves this problem by providing an AI-powered, web-based platform where users can upload any dataset, which is then automatically preprocessed, analyzed, and visualized. The system detects target columns, selects suitable models, generates KPIs, and allows natural language queries that are translated into SQL\/Pandas, enabling quick, accurate, and accessible decision-making.","description":"Our FYP builds an AI-powered, web-based data analytics assistant that allows users to upload any CSV or Excel dataset, WHICH WILL FIRST GO THROUGH AN AUTOMATED PREPROCESSING PYTHON PIPELINE THAT HANDLES MISSING VALUES, OUTLIERS, AND FEATURE ENGINEERING.\r\nAfter preprocessing, THE SYSTEM AUTOMATICALLY DETECTS WHETHER THE DATASET IS FOR CLASSIFICATION, REGRESSION, OR UNSUPERVISED LEARNING BY ANALYZING COLUMN TYPES, UNIQUE VALUES, AND TARGET PREDICTABILITY. If the target column is categorical with limited unique values, IT USES CLASSIFICATION MODELS (E.G., LOGISTIC REGRESSION, DECISION TREE, RANDOM FOREST). IF THE TARGET IS NUMERIC, REGRESSION MODELS ARE APPLIED (E.G., LINEAR REGRESSION, RANDOM FOREST REGRESSOR, GRADIENT BOOSTING). IF NO SUITABLE TARGET IS FOUND, THE SYSTEM APPLIES UNSUPERVISED MODELS (E.G., K-MEANS CLUSTERING).\r\nUSERS CAN ENTER NATURAL LANGUAGE QUERIES, WHICH ARE FIRST PREPROCESSED THROUGH A PYTHON PIPELINE (TOKENIZATION, LOWERCASING, STOP-WORD REMOVAL, ETC.) and then translated into SQL\/PANDAS commands via an LLM (such as GPT or Gemini) and executed on the dataset. The system then generates KPIs, predictions, and trends, and automatically recommends suitable visualizations (line, bar, pie, scatter), displayed in an interactive dashboard for fast, non-technical insight generation. The system provides future recommendation by predicting future outcomes and generating data-driven suggestion to support better decision making.$$\n1. Dataset Preprocessing Pipeline \r\nCleans and prepares datasets by handling missing values, outliers, and categorical encodings. Example: If 20% of \u201cProfit\u201d values are missing, they are replaced with the median. \r\n \r\n2. Target Column & Model Selection \r\nAutomatically detects problem type (classification, regression, clustering, forecasting) and selects suitable ML models. \r\nExample: If column \u201cChurn\u201d = Yes\/No \u2192 classification, if \u201cRevenue\u201d is numeric \u2192 regression. \r\n \r\n3. Natural Language Query Engine \r\nAllows users to ask questions in plain English, which are converted to SQL\/Pandas queries. \r\nExample: Query: \u201cShow top 5 products by sales\u201d \u2192 SQL executed \u2192 Bar chart generated. \r\n \r\n4. Machine Learning Module \r\nApplies ML algorithms to discover insights and predict trends. Example: Forecast predicts 12% sales growth in Q1 2026. \r\n\u2022\tThis module is divided into to sub modules\r\n ->Supervised learning Model \r\n ->Unsupervised learning Model\r\n\r\n5. Visualization & Dashboard Module \r\nInteractive dashboards with multiple persistent visualizations and KPIs. \r\nExample: Users can view sales trend, top 10 products, and profit by region side by side. \r\n\r\n6. Database Module\r\nStores processed datasets, model results, and user queries dynamically using PostgreSQL \/ MongoDB for efficient retrieval.\r\nExample: After a user uploads sales data, it is stored in PostgreSQL. When the user later queries \u201cTop customers by profit,\u201d results are fetched directly from the database.$$\n1.\tTarget Column & Model Selection Module\r\n2.\tVisualization & Dashboard Module \r\n3.\tDeveloped supervised learning Model Module (e.g:Linear Regression,Decision Tree, Random Forest etc).$$\n1.\tDataset Preprocessing Pipeline \r\n2.\tNatural Language Query Engine Module \r\n3.\tDatabase (PostgreSQL\/MongoDB)\r\n4.\tDeveloped unsupervised learning Model Module(e.g:K-Means Clustering, DBSCAN, Apriori, FP-Growth etc).$$\n$$\n$$\n$$\n$$\n","comments":" $$ Ensure implementation of multiple AI model as sated in revised proposal.","isDraft":0,"status":2,"created_at":"2025-10-01 16:12:18","updated_at":"2025-10-08 15:40:26","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1023,"project_id":1367,"title":"Time Explorer-An engaging app where children discover history through fun, colorful adventures.","prob":"Students and young learners often perceive history as a dry or challenging subject when taught through traditional methods such as textbooks, lectures, and rote memorization. This lack of engagement reduces their interest and prevents them from truly connecting with historical events, figures, and cultures.\r\n\r\nTo address this problem, our project introduces a gamified and interactive mobile application designed specifically for kids. The app transforms history into an exciting adventure by allowing learners to explore ancient civilizations, interact with famous historical figures, explore important landmarks and events. Through games, quizzes, animations, and storytelling, history becomes fun, engaging, and easy to understand. This approach not only builds curiosity but also strengthens knowledge retention by turning learning into an enjoyable experience.","description":"Time Explorer :\r\n(i) Historical Places Explorer Module\r\nAllows learners to explore 50 historical landmarks from Asia and Europe through images, maps, and videos.\r\nPlaces are categorized into Ancient Wonders, Cultural Heritage, Architectural Landmarks, Battlefields, and Religious Sites, with data from Wikipedia, UNESCO, Britannica, and Google Arts & Culture, structured in JSON for AI integration.\r\n\r\n(ii) Meeting with Historical Figures Module\r\nEnables virtual interaction with 50 historical personalities from Asia and Europe, categorized as Scientists, Physicists, Philosophers, Leaders, and Artists.\r\nUsing Gemini AI and prompt engineering, learners can simulate conversations powered by JSON knowledge bases from verified sources.\r\n\r\n(iii) Quizzes & Challenges Module\r\nOffers interactive quizzes and challenges on places and figures with AI-based difficulty adjustment.\r\nTracks performance to reinforce learning and identify strengths or weak areas.\r\n\r\n(iv) Progress Tracking Module\r\nMaintains a personalized learning profile, recording achievements, badges, and completed activities.\r\nDisplays visual progress charts and reports for both learners and admins.\r\n\r\n(v) SQLite Local Storage & Firebase Sync Module\r\nEnsures offline and online continuity by storing data locally in SQLite and syncing with Firebase when connected.\r\nPrevents duplication through automated cleanup after sync.\r\n\r\n(vi) Admin Dashboard Module\r\nAllows admins to manage content, quizzes, and AI datasets, view user reports, and refine prompts.\r\nEnsures data remains accurate, updated, and AI-ready for smooth operation.\r\n\r\n(vii) Animated & Engaging Front-End Module\r\nDelivers a modern, interactive UI using Flutter\u2019s Material 3 and animation libraries.\r\nProvides smooth navigation and visual effects, making learning immersive and enjoyable.$$\n(i) Historical Places Explorer Module\r\nAllows learners to explore 50 historical landmarks from Asia and Europe through images, maps, and videos.\r\nPlaces are categorized into Ancient Wonders, Cultural Heritage, Architectural Landmarks, Battlefields, and Religious Sites, with data from Wikipedia, UNESCO, Britannica, and Google Arts & Culture, structured in JSON for AI integration.\r\n\r\n(ii) Meeting with Historical Figures Module\r\nEnables virtual interaction with 50 historical personalities from Asia and Europe, categorized as Scientists, Physicists, Philosophers, Leaders, and Artists.\r\nUsing Gemini AI and prompt engineering, learners can simulate conversations powered by JSON knowledge bases from verified sources.\r\n\r\n(iii) Quizzes & Challenges Module\r\nOffers interactive quizzes and challenges on places and figures with AI-based difficulty adjustment.\r\nTracks performance to reinforce learning and identify strengths or weak areas.\r\n\r\n(iv) Progress Tracking Module\r\nMaintains a personalized learning profile, recording achievements, badges, and completed activities.\r\nDisplays visual progress charts and reports for both learners and admins.\r\n\r\n(v) SQLite Local Storage & Firebase Sync Module\r\nEnsures offline and online continuity by storing data locally in SQLite and syncing with Firebase when connected.\r\nPrevents duplication through automated cleanup after sync.\r\n\r\n(vi) Admin Dashboard Module\r\nAllows admins to manage content, quizzes, and AI datasets, view user reports, and refine prompts.\r\nEnsures data remains accurate, updated, and AI-ready for smooth operation.\r\n\r\n(vii) Animated & Engaging Front-End Module\r\nDelivers a modern, interactive UI using Flutter\u2019s Material 3 and animation libraries.\r\nProvides smooth navigation and visual effects, making learning immersive and enjoyable.$$\nIn the Time explorer Application, Usman Ayaz shall develop the frontend of the app along with the following modules:\r\n\r\n(i) Historical Places Explorer Module\r\n\r\n(ii)Animated & Engaging Front-End Module\r\n\r\n(iii) SQLite Local Storage & Firebase Sync Module\r\n\r\n(iv) Admin Dashboard Module\r\n\r\nBackend & APIs Integration.$$\nIn the Time explorer Application, Abdul Moeed shall develop the AI-based modules, which include:\r\n(i) Meeting with Historical Figures Module (AI Chat)\r\n\r\n(ii) Quizzes & Challenges Module (AI-Generated Quizzes)\r\n\r\n(iii) Progress Tracking Module\r\n\r\n(iv) AI Processing & Data Enhancement$$\n$$\n$$\n$$\n$$\n","comments":" $$ The revised Time Explorer proposal demonstrates a substantial improvement in clarity, technical scope, and feasibility. The team has effectively addressed previous concerns regarding the static database, unclear scope, and weak understanding by defining dynamic data sources, structured JSON integration, and AI-powered modules using Gemini AI. The inclusion of offline\u2013online synchronization, animated front-end design, and well-defined user roles makes the application both educational and engaging for young learners. The project now reflects strong conceptual maturity and technical planning, with only minor clarification needed regarding AI data updates and content moderation. Overall, the revision is comprehensive, innovative, and ready for approval.","isDraft":0,"status":2,"created_at":"2025-10-13 12:22:18","updated_at":"2025-10-17 08:56:48","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1042,"project_id":1378,"title":"Seerah Timeline","prob":"The Seerah Timeline App addresses the challenge of limited access to structured and engaging resources on the life of Prophet Muhammad (SAW), as traditional books and scattered content often overwhelm learners, making it difficult to follow events, retain lessons, and apply them in daily life. The app offers an interactive timeline of the Prophet\u2019s life from birth to demise with authentic references from Ar-Raheequl Makhtum, and also add more books releated with seerah, concise summaries, miracles, event and lessons. To make learning more effective, a RAG-based AI chatbot is integrated, which is trained on our own dataset of authentic Islamic books provided in PDF format. This chatbot retrieves information directly from the uploaded sources we add in app and generates accurate, context-based answers to user queries along with concise summaries, ensuring that learners receive reliable, personalized, and easily accessible Islamic knowledge that promotes both spiritual growth and practical application.","description":"The Seerah Timeline App is an interactive mobile application designed to present the life of Prophet Muhammad \ufdfa in a structured, chronological, and engaging format. It provides users with a comprehensive timeline covering major events from the Pre-Prophethood, Makkan, and Madinan eras, enriched with authentic references from Ar-Raheequl Makhtum, detailed descriptions, concise summaries, lessons, miracles, and multimedia support such as images, narrations, videos (from YouTube), and interactive maps. Users can bookmark favorite topics, receive personalized notifications like \u201cEvent of the Day,\u201d and test their knowledge through quizzes and interactive learning modules. To further enhance learning, the app integrates a RAG-based AI chatbot, trained on our curated dataset of authentic Islamic books provided in PDF format. This chatbot intelligently retrieves relevant information from the dataset and generates accurate, context-based answers and summaries to user queries. By combining structured timelines, interactive learning, and AI-powered assistance, the Seerah Timeline App offers a unique blend of education, spirituality, and technology, making the Seerah of Prophet Muhammad \ufdfa accessible, relevant, and practically applicable in daily life.$$\nThe Seerah Timeline App is designed with multiple modules that collectively provide an engaging and educational experience. The User Authentication & Profile Module manages signup, login, and guest access while allowing customization options such as language, theme, font size, and notifications for a personalized experience. The Seerah Timeline Module presents the life of Prophet Muhammad \ufdfa in a chronological format divided into the Pre-Prophethood, Makkan, and Madinan eras, with each entry linking to detailed information for structured navigation. The Event Details & Lessons Module offers authentic references, comprehensive descriptions, related media like images and videos. To reinforce learning, the Quiz & Interactive Learning Module provides multiple-choice quizzes. The Bookmarks & Favorites Module enables users to save important events, topics, and lessons for quick access, while the Notifications & Daily Content Module ensures continuous engagement with reminders about significant events, daily Seerah lessons, and \u201cEvent of the Day. Finally, the AI Chatbot Module (RAG-Based), trained on a curated dataset of authentic Islamic books in PDF format, uses Retrieval-Augmented Generation (RAG) to fetch relevant information and deliver accurate, context-based answers and concise summaries, ensuring personalized, reliable, and interactive knowledge acquisition that further enhances the overall learning experience.$$\nThe User Authentication & Profile Module manages signup, login, and guest access, offering customizable settings such as language, theme, font, and notifications. The Seerah Timeline Module presents the life of Prophet Muhammad \ufdfa in chronological order, divided into Pre-Prophethood, Makkan, and Madinan eras. The Event Details & Lessons Module provides event descriptions, authentic references, media, and highlights morals and rulings for practical guidance.$$\nThe Quiz & Interactive Learning Module provides multiple-choice quizzes, Q&A, and score tracking to make learning engaging through gamification. The Bookmarks & Favorites Module allows users to save important events and lessons, creating a personalized library for quick access. The Notifications & Daily Content Module sends daily or weekly reminders, such as \u201cEvent of the Day\u201d or short Seerah lessons, ensuring regular user engagement. The AI Chatbot Module (RAG-Based), trained on a curated dataset of authentic Islamic books in PDF format, uses Retrieval-Augmented Generation (RAG) to fetch relevant information and deliver accurate, context-based answers and concise summaries, ensuring personalized, reliable, and interactive knowledge acquisition that further enhances the overall learning experience.$$\n$$\nSeerah of Prophet Muhammad \ufdfa . Such apps are not being developed in our department. This comparison is in contrast with external seerah app.$$\nAI Chatbot: Gives answers from Islamic books using RAG technology.$$\nQuizzes: Has quizzes to test knowledge$$\nDaily Reminders: Sends notifications like \u201cEvent of the Day\u201d.","comments":" $$ The integration of a RAG-based chatbot, quizzes, and interactive learning tools makes it innovative and educationally impactful. While minor gaps remain in describing the AI architecture and dataset flow, $$ Take care about these minor concerns","isDraft":0,"status":2,"created_at":"2025-10-23 13:03:28","updated_at":"2025-10-27 14:35:04","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":974,"project_id":1351,"title":"VocalID: Intelligent Voice Authentication with Spoof & Liveness Detection","prob":"Traditional authentication methods like passwords and PINs are prone to theft, phishing, and user error. Although biometrics such as fingerprints and facial recognition provide stronger security, they can be spoofed with high-resolution images or silicone molds and require specialized hardware, leaving systems vulnerable. Voice biometrics offers a convenient, contactless, and hardware-agnostic alternative but faces challenges from sophisticated spoofing attacks, including replay attacks with pre-recorded voice clips and AI-generated deepfakes.\r\nThis project addresses these security risks by developing a secure, passwordless voice authentication system that verifies the user is a live, genuine human at the time of access. By effectively detecting and blocking replay and AI-based spoofing attacks, it enhances security for critical applications like banking, corporate access, and personal devices. This approach ensures safer, more reliable voice biometric authentication, overcoming the limitations of traditional and biometric methods.","description":"'VocalID' is a cross-platform voice authentication system. A manager enrolls users by recording their voice to create a unique voiceprint. For login, the system displays a random alphanumeric phrase (e.g., \"blue42fox\"). The user must speak this phrase aloud. Their audio is processed through a multi-stage security pipeline: first, the spoken words are verified against the on-screen challenge to prove liveness. Second, the voice's characteristics are matched against their enrolled voiceprint. Crucially, a dedicated AI subsystem analyzes the audio in real-time to detect spoofing attempts, such as replays or synthetic voices, using a fine-tuned model. This combined approach of dynamic challenge-response and advanced spoof detection ensures robust security. The system provides a consistent, hardware-free login experience on both web and mobile platforms, offering a convenient yet highly secure alternative to traditional passwords and other biometrics.$$\n1. User Management Module: Handles all user-related functionalities. This includes the registration system for both end-users and managers, a secure voice enrollment interface for creating initial voiceprints, implementation of Role-Based Access Control (RBAC) to manage permissions, and a dashboard for profile management.\r\n\r\n2. AI Spoof Detection System: The core security layer against attacks. This module involves fine-tuning a pre-trained model (e.g., AASIST) on datasets like ASVspoof. It performs audio feature extraction (e.g., LFCCs) and classifies the input audio as genuine or a spoof, protecting against replay, text-to-speech, and voice conversion attacks.\r\n\r\n3. Random Phrase Generation Module: Responsible for proving user liveness. It generates dynamic, random, and pronounceable alphanumeric strings for each authentication attempt. It securely serves these challenges to the frontend and manages their lifecycle (generation, display, and expiration).\r\n\r\n4. Voice Recording & Processing Module: Manages the audio input. It provides a cross-platform interface for recording audio in the user's browser and mobile app. It then preprocesses the raw audio by performing noise reduction, silence trimming, and format conversion, and extracts relevant features (e.g., MFCCs) for further analysis.\r\n\r\n5. Voice Matching & Liveness Verification Module: Makes the final authentication decision. It verifies that the words spoken by the user exactly match the generated challenge phrase (text-dependent verification). It then compares the live voice's characteristics against the stored user voiceprint. Based on the results of this check and the spoof detection system, it grants or denies access.$$\nI will be the primary developer for the User Management Module (backend API and frontend interfaces) and the Random Phrase Generation Module (algorithm and integration). I will also co-develop the Voice Recording & Processing Module, focusing on building the recording interface and implementing basic audio preprocessing steps to prepare the audio for the AI and matching subsystems.$$\nI will be the primary developer for the AI Spoof Detection System (fine-tuning the model on spoof datasets and integration) and the Voice Matching & Liveness Verification Module (developing the matching logic). I will also co-develop the Voice Recording & Processing Module, with a focus on implementing the advanced audio feature extraction (e.g., MFCCs) required for both the spoof detector and the voice matching algorithm.$$\n$$\nN\/A$$\nFine-Tuned AI Spoof Detection Model: Developing and fine-tuning an anti-spoofing model (e.g., AASIST) on audio datasets to detect advanced deepfakes and synthetic audio missed by standard models.$$\nDynamic Challenge-Response Liveness Verification: System generates random pronounceable alphanumeric phrases for each login, requiring users to repeat them to confirm live presence and block replays.$$\nCross-Platform Voice Authentication Pipeline: A unified system ensuring the same security and user experience for spoof detection and liveness checks across web browsers and Android\/iOS platforms.","comments":" $$ student have provided the detailed responsibility of the modules. System should work as per description and the main task (live human voice detection)","isDraft":0,"status":2,"created_at":"2025-10-03 10:40:23","updated_at":"2025-10-15 15:20:39","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1013,"project_id":1366,"title":"Dynamic Reinforcement Learning Firewall with Adaptive Threat Detection","prob":"Traditional firewalls rely on static rules and signature-based detection, making them ineffective against zero-day attacks, advanced persistent threats, and evolving malware. Current network security solutions lack adaptive learning capabilities to automatically adjust protection strategies based on emerging threat patterns. This creates security gaps where new attack vectors can bypass conventional defenses, leaving networks vulnerable to sophisticated cyber attacks that cost organizations billions annually.","description":"This project develops an intelligent firewall system using reinforcement learning algorithms that dynamically adapts its security policies based on real-time network traffic analysis. The system learns from attack patterns, automatically updates firewall rules, and provides predictive threat detection capabilities.\r\n\r\nIt features a web-based dashboard for monitoring, rule management, and performance analytics, making network security proactive rather than reactive.\r\n\r\nThe system integrates Deep Q-learning networks with traditional packet filtering to create an adaptive security layer that evolves with emerging threats. It includes real-time traffic analysis, automated rule generation, threat intelligence integration, and comprehensive logging for forensic analysis.$$\n1. Reinforcement Learning Engine Module\r\nCore AI component implementing Deep Q-Network (DQN) and Policy Gradient algorithms. Processes network traffic features, maintains reward mechanisms for correct threat classifications, and continuously updates decision-making models. Includes performance optimization, model persistence, and integration with traffic analysis components for real-time threat classification.\r\n1.1. Raw Sockets: The system will use RAW sockets to capture network packets directly from the network layer. This allows packet exploration before the IP stack for real-time inspection.\r\n1.2. New day attack: To handle zero-day (new-day) attacks, the system will focus on anomaly detection \u2014 identifying traffic patterns that differ from known behaviour and updating policies automatically through reinforcement learning.\r\n1.3. Datasets: The model will be trained using public intrusion detection datasets such as CICIDS2017, NSL-KDD, or UNSW-NB15 for realistic attack scenarios.\r\n2. Traffic Analysis & Feature Extraction Module\r\nReal-time packet capture and analysis. Extracts features such as packet size, protocol types, connection patterns, payload characteristics, and temporal behaviors. Implements statistical analysis and anomaly detection algorithms.\r\n3. Dynamic Rule Management System\r\nAutomatically generates, updates, and manages firewall rules based on RL model decisions. Maintains rule versioning, rollback capabilities, and conflict resolution mechanisms. Provides API integration with existing firewall hardware\/software.\r\n4. Web-based Dashboard & Monitoring Interface\r\nA web-based dashboard will be created using Streamlit or Flask, showing live traffic monitoring, detected attacks, model accuracy, and firewall rule updates. This dashboard will be used for project demos during the Open House and evaluations.\r\n5. Features: \r\nLive Packet Capture: Captures and analyzes real-time network traffic using RAW sockets.\r\nDynamic Rules: Updates firewall rules automatically through learning.\r\nRL Engine (DQN): Learns from traffic patterns to allow or block packets.\r\nAnomaly Detection: Finds unusual or unknown attack behaviors.\r\nDashboard: Shows live traffic, detected attacks, and system status.$$\nModules: Traffic Analysis & Feature Extraction, Dynamic Rule Management System\r\nTasks: Packet capture mechanisms, feature engineering pipelines, statistical analysis, anomaly detection, and automated rule generation with conflict resolution and rollback mechanisms.$$\nModule: Reinforcement Learning Engine,Web-based Dashboard & Monitoring Interface\r\nTasks: Implement DQN algorithms, reward systems, model training pipelines, and AI decision-making logic. Includes performance optimization, model persistence, and real-time threat classification mechanisms.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Students have updated the proposal as per the comitee comments. The modules must be developed as claimed in the proposal","isDraft":0,"status":2,"created_at":"2025-10-08 21:24:39","updated_at":"2025-10-15 15:26:45","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1029,"project_id":1383,"title":"EmoTales AI: Emotion-Aware Educational Story & Comic Generator","prob":"Modern education faces a lack of emotionally engaging learning content. Teachers and parents often struggle to design personalized moral or educational stories that hold students\u2019 attention. Existing AI tools can only caption images or generate generic stories without emotional depth or educational focus.\r\nOur project, EmoTales AI, aims to bridge this gap by introducing an AI-powered system that transforms images or short text prompts into emotion-aware educational stories and illustrated comics. Using our custom-trained CNN-based emotion detection module, the system adapts story tone and visuals according to the detected emotions of characters. It generates coherent, personalized narratives suitable for students, parents, and teachers.\r\nThis will help make education interactive, moral-based, and emotionally intelligent, encouraging children to connect emotionally with what they learn.","description":"EmoTales AI is an innovative web-based educational storytelling platform powered by Artificial Intelligence that supports both Image-to-Story and Text-to-Story generation modes.\r\n\r\nIn Image-to-Story mode, users upload multiple educational or real-life images. The system automatically detects faces, classifies their emotions using a custom-trained CNN model built using the FER-2013 dataset and an additional self-trained images dataset, and identifies key objects and characters through YOLOv8 and BLIP-2. This extracted information is then processed by a story generation model that produces a coherent, emotion-aware story maintaining narrative flow and character consistency across all uploaded images.\r\n\r\nIn Text-to-Story mode, users provide a brief idea or topic such as honesty, teamwork, or helping others. The system utilizes a Large Language Model (LLM) to expand it into a structured, moral-based educational story. Each generated story is enriched with AI-generated illustrations using Stable Diffusion and ControlNet, and accompanied by text-to-speech narration for auditory learners.\r\n\r\nThe platform also provides multilingual translation, emotion-based tone control, and export options such as illustrated PDFs and audio files. The complete system will be developed using React.js for the frontend, Node.js and Flask (Python) for the backend, and MongoDB\/MySQL for efficient data management.\r\n\r\nBy combining AI, emotion recognition, and storytelling, EmoTales AI makes education emotionally intelligent, interactive, and creative \u2014 helping students connect moral lessons with emotional understanding while improving comprehension and engagement in learning.$$\nThe proposed system, EmoTales AI, consists of five integrated modules that collectively perform image understanding, story generation, emotion analysis, comic illustration, and presentation through a web interface. Each module interacts through backend APIs to ensure smooth and consistent workflow.\r\n\r\n1\ufe0f\u20e3 Image Understanding Module\r\n\r\nThis module performs all image analysis and preprocessing tasks. It enables multi-image upload and ordering through an interactive drag-and-drop interface, allowing users to maintain story sequence visually. Each image is analyzed to extract meaningful content using a combination of AI-based techniques.\r\n\r\nEmotion detection is carried out using a custom-trained CNN model trained on the FER-2013 dataset and additional self-trained images dataset, capable of identifying emotions such as happiness, sadness, anger, fear, and surprise. Detected emotions are used to add emotional depth to the generated narrative.\r\nAdditionally, YOLOv8 and OpenCV are utilized for object and character detection, while BLIP-2 is employed to generate descriptive captions and keywords. The combined output (emotion labels, objects, and captions) forms the contextual foundation for narrative generation.\r\n\r\n2\ufe0f\u20e3 Story Generation & Presentation Module\r\n\r\nThis module generates a coherent educational story based on the input images and extracted contextual data. A Large Language Model (LLM) such as GPT or LLaMA processes keywords, objects, and emotions to create a story that aligns with the educational objective.\r\nThe generated story adapts tone and style dynamically (e.g., happy, funny, dramatic, or moral). It also produces titles, summaries, and lesson highlights. Story presentation is handled through a web-based viewer that allows users to explore scene-by-scene narratives.\r\nThis module also manages translation (English, Urdu, etc.) and supports exporting the story in multiple formats, including text, comic layout, and audio narration.\r\n\r\n3\ufe0f\u20e3 Comic & Narration Module\r\n\r\nThis module transforms generated stories into visual comic scenes. It employs Stable Diffusion and ControlNet for image generation, ensuring alignment between the story text, character emotions, and visual context. Facial and body expressions are adapted to match the detected emotions for more expressive visuals.\r\nFor narration, the system integrates Google Cloud Text-to-Speech (TTS) to provide natural audio playback. Users can listen to the story, improving accessibility and engagement. The final story can be exported as text, illustrated PDF, or audio format.\r\n\r\n4\ufe0f\u20e3 Integration & Backend Module\r\n\r\nThis module connects all AI models and ensures communication between components. Built using Node.js and Flask (Python) microservices, it manages data flow between modules such as emotion detection, story generation, and illustration.\r\nIt handles model requests, API endpoints, and ensures smooth execution of hybrid AI tasks (pre-trained + custom models). Additionally, it manages storage and retrieval operations through MongoDB\/MySQL, maintaining logs of generated stories and user sessions.\r\n\r\n5\ufe0f\u20e3 Web Interface Module\r\n\r\nThe user-facing interface is developed in React.js for an interactive, intuitive experience. It includes options for text input, image uploads, story style selection, and real-time preview. The story viewer and comic preview panel enable users to view generated content scene by scene.\r\nThis module ensures a clean, educational, and emotionally engaging storytelling experience accessible from any web browser.$$\nI will develop the Story Generation, Emotion Detection, and Export Modules, focusing on narrative logic, AI-based emotion adaptation, and story presentation.\r\n\r\nResponsibilities:\r\n\r\nIntegrate Large Language Models (LLMs) for Image-to-Story and Text-to-Story generation.\r\n\r\nImplement story logic with educational and moral tone adjustments.\r\n\r\nDevelop emotion detection API integration using the custom CNN module.\r\n\r\nManage Google Cloud TTS and AI Narrator Voice Personality (child, teacher, storyteller tones).\r\n\r\nImplement multilingual translation (English, Urdu).\r\n\r\nHandle export features for stories (PDF, comic, and audio).\r\n\r\nTest emotion-to-narration consistency and overall narrative performance.$$\nI will develop the Image Understanding, Comic Visualization, and System Integration Modules, focusing on deep learning, visual storytelling, and inter-module communication.\r\n\r\nResponsibilities:\r\n\r\nCreate multi-image upload and ordering interface (drag-and-drop UI).\r\n\r\nImplement object and character detection using YOLOv8 and OpenCV.\r\n\r\nIntegrate BLIP-2 for caption and keyword extraction.\r\n\r\nGenerate AI-based comic illustrations using Stable Diffusion + ControlNet.\r\n\r\nMaintain emotion\u2013visual alignment across comic scenes.\r\n\r\nImplement system integration and data flow management using Flask and Node.js.\r\n\r\nDesign the Story Viewer for comic visualization and scene-wise navigation.$$\n$$\nNULL$$\n1) Custom-Trained Emotion Detection Model (CNN)$$\n2)Emotion-Aware Educational Story Generation$$\n3)AI-Generated Comic Visualization with Translation and Narration","comments":" $$ Ensure are all the changes incorporated in the revised proposal","isDraft":0,"status":2,"created_at":"2025-10-16 21:53:34","updated_at":"2025-10-27 14:36:57","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1036,"project_id":1333,"title":"Cogniscan: AI-powered App for early dyslexia and dysgraphia detection","prob":"Students with learning difficulties such as dyslexia and dysgraphia often remain undiagnosed due to lack of specialized resources in Pakistan. Without early diagnosis and personalized support, they suffer from poor academic performance, low self-esteem, and emotional stress. Current educational tools are limited to assessments only and fail to provide targeted therapy or cognitive exercises.\r\nCogniScan addresses this gap by providing a mobile-based platform that not only detects dyslexia and dysgraphia through AI-powered handwriting analysis but also recommends personalized therapy exercises using machine learning. The app is designed for neurodivergent children, their parents, and psychiatrists to collaboratively monitor progress and apply timely interventions. By making it mobile-based, it increases accessibility and real-time interaction, especially for students in underserved regions.","description":"CogniScan is an AI-powered mobile application developed to detect early signs of dyslexia and dysgraphia through automated handwriting analysis. These learning disorders often remain undiagnosed in early school years due to limited awareness and lack of screening tools, which leads to academic struggle and low confidence among children. CogniScan provides an affordable and intelligent screening solution that identifies writing-based learning challenges and supports early intervention through AI-based assessment and personalized learning support.\r\n\r\nFor AI training, CogniScan will utilize two research-based handwriting datasets:\r\n\r\n1.Potential Dysgraphia Handwriting Dataset of School-Age Children (Mendely Data)\r\n2.Dyslexia Handwriting Dataset (Kaggle)\r\n\r\nThese datasets contain labeled handwriting samples that help train the AI model to classify and differentiate between normal handwriting patterns and disorder-related writing behaviors.\r\n\r\nInference of Personalized Exercises Using Threshold-Based AI Approach:\r\nOnce handwriting is analyzed, the AI classification model generates a probability score between 0 and 1 to indicate the likelihood of a learning disorder. A threshold-based mechanism is applied to categorize severity levels:\r\nNo difficulty detected\r\nMild difficulty\r\nModerate difficulty\r\nSevere difficulty\r\nBased on the detected severity, a second AI model is used for exercise recommendation. This model uses a Decision Tree-based recommendation system trained on handwriting error patterns and learning progress data. . The system follows adaptive learning, meaning it adjusts exercise difficulty based on user performance over time. This two-phase AI approach ensures accurate diagnosis and effective personalized learning support within the CogniScan mobile application.$$\n1. User & Role Management\r\nSelf-registration for students, parents, and psychiatrists.\r\nRole-based access control.\r\nChild-parent linking system.\r\n\r\n2. AI-based Diagnostic Module\r\nInitial test, dyslexia test, and dysgraphia test.\r\nAI handwriting analysis via image\/text input.\r\nInstant results with severity level indicators.\r\n\r\n3. Exercise Recommendation System (New Core Module)\r\nAI-based recommendation model using a Decision Tree classifier trained on handwriting error patterns from the Dyslexia Handwriting Dataset and Potential Dysgraphia Handwriting Dataset.\r\n\r\n4. Progress Tracking & Dynamic Plan Adjustment\r\nRegular retesting and progress input.\r\nExercise adaptation via retraining or rule-based logic.\r\nVisual analytics for parents and psychiatrists.\r\n\r\n5. Appointment & Consultation Module\r\nParents can schedule appointments with psychiatrists.\r\nPsychiatrists can update treatment notes and feedback.\r\n\r\n6. Secure Payment Gateway\r\nBookings for psychiatrist sessions.\r\nIn-app purchases for advanced features.\r\n\r\n7. Admin Panel\r\nApproves psychiatrists.\r\nManages users, exercises, tests, and reports.$$\nI shall develop:\r\n\r\nPsychiatrist Registration & Admin Approval\r\nPsychiatrists submit their credentials for verification. Admins validate and approve profiles to ensure licensed professionals are on the platform.\r\n\r\nAppointment & Consultation System\r\nParents can book appointments, view psychiatrist profiles, and receive session notes. Psychiatrists can manage consultations and update progress.\r\n\r\n Payment Integration\r\nSecure in-app payments using Stripe or similar API for consultation and premium services. Generates invoices and stores payment logs.\r\n\r\nAdmin Dashboard\r\nManages users, test data, exercise content, payments, and psychiatrist approval workflow.\r\n\r\nAnalytics & Reports\r\nGenerates visual reports for user progress, test results, and therapy outcomes using charts and summary dashboards.\r\n\r\n AI-Powered Handwriting Analysis\r\nThis module processes handwriting samples (uploaded as images) to detect early signs of dysgraphia using a machine learning pipeline trained on handwriting features. The results are used to generate diagnostic scores and reports.$$\nI shall develop:\r\nUser Registration and Role Management\r\nAllows students, parents, and psychiatrists to sign up. Parents can manage their child\u2019s profile and link accounts.\r\n\r\nInitial Assessments \r\nEnables users to attempt dysgraphia screening tests and assessments based on standard cognitive evaluation methods.\r\n\r\nML-Based Exercise Recommendation System (Core Feature)\r\nTrains a model to suggest appropriate cognitive and motor-skill-based writing exercises. The model adapts based on user data, difficulty levels, and improvement history.\r\n\r\nLearning Dashboard\r\nDisplays exercises, and progress.\r\n\r\nProgress Tracker & Re-Assessment Tools\r\nVisualizes improvement over time, allows retesting, and syncs data with psychiatrists and parents for real-time review.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Ensure all changes incorporated in the revised version.","isDraft":0,"status":2,"created_at":"2025-10-20 19:09:06","updated_at":"2025-10-27 14:29:11","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":984,"project_id":1369,"title":"Manzil - Smart Career Counseling Portal","prob":"Pakistani students face major confusion after completing intermediate (FSc, FA, ICS, ICOM)\r\ndue to the absence of professional career counseling and a centralized educational guidance\r\nsystem. Many select degrees without understanding their true interests, job market scope,\r\nfinancial feasibility, or university options. This often results in wrong career choices, wasted\r\ntime and money, and unemployment. Manzil aims to solve this problem by providing a web based platform offering AI-powered career counseling, university admissions information, entry test guidance, last year\u2019s merit lists, scholarships, and financial aid details, all in one\r\nplace.","description":"Manzil is a comprehensive web-based (Industrial Project) career counseling and university guidance portal designed to help students make informed academic choices. Students can create profiles, attempt three standardized career tests assessing their personality, aptitude, and interests, and receive personalized recommendations for suitable degrees, careers, and universities. Unlike existing platforms such as MeraFuture, Manzil uses a weighted aggregation model to combine test results, ensuring balanced and consistent recommendations rather than relying on a single assessment.\r\n\r\nThe system gathers authentic and up-to-date information about HEC-recognized universities, including admission deadlines, eligibility criteria, and last year\u2019s merit lists. Data is collected through web scraping of official university websites, open APIs (where available), and manual verification when scraping is not feasible. It also centralizes entry test guidance, scholarship opportunities, and integrates validated career aptitude datasets such as the Holland Codes.\r\n\r\nA hybrid data analysis approach, combining rule-based mappings (e.g., logical + analytical \u2192 Engineering; creative + communication \u2192 Media) with dataset-driven matching to student profiles, ensures accurate recommendations. The system flow follows a structured process: (1) students attempt three tests, (2) results are normalized and aggregated using score mechanism, (3) profiles are matched with career fields and university criteria, and (4) ranked recommendations are generated. By integrating career counseling, admissions data, and financial aid information in one intelligent system, Manzil reduces confusion and empowers students to confidently choose the right academic and professional path.$$\n1) Student Profile & Dashboard\r\nThis module allows students to create accounts, enter their academic background, and manage preferences. The dashboard serves as a personalized hub, displaying career suggestions, admission deadlines, test schedules, and notifications for quick access.\r\n\r\n2) Career Counseling Module (Aptitude & Interest Test)\r\nStudents take aptitude and interest-based assessments, and the system recommends suitable careers and degrees. AI-driven analysis ensures alignment between student skills, interests, and future opportunities.\r\n\r\n3) Degree & Career Scope (Job Market, Future Trends)\r\nThis module provides insights into the scope of degrees, job opportunities, salary expectations, and industry growth trends. It helps students evaluate long-term career viability before making decisions.\r\n\r\n4) University Information Module\r\nA centralized database of HEC-recognized universities offering details about programs, fee structures, facilities, rankings, and campus locations. Students can easily explore universities and compare options.\r\n\r\n5) Admissions & Testing Module\r\nThis provides admission deadlines, eligibility criteria, and direct application links. It also integrates entry test information (NTS, ECAT, MDCAT, USAT, etc.) and includes last year\u2019s merit lists to help students assess their chances realistically.\r\n\r\n6) AI Chatbot Career Counselor\r\nAn NLP-powered chatbot that answers student queries instantly. It provides guidance on admissions, career scope, scholarships, and test details in real time, reducing manual searching.\r\n\r\n7) University & Degree Recommendation System\r\nBased on student profiles, test results, and preferences, this module suggests suitable universities and programs. Recommendations are personalized, considering both academic strengths and career goals.\r\n\r\n8) Comparison Tool\r\nStudents can compare two or three universities or degree programs side by side. Comparisons cover fees, merit criteria, facilities, rankings, and scope, helping students make well-informed choices.\r\n\r\n9) Notification Alerts\r\nThis ensures students never miss important deadlines or opportunities. Alerts include admission dates, test schedules, scholarship announcements, and merit list updates, delivered via in-app, email, or SMS notifications.\r\n\r\n10) Student Community & Discussion Forum\r\nA peer-to-peer platform where students interact, ask questions, and share experiences. It supports discussions on admissions, hostel facilities, accommodation, and test preparation, fostering a collaborative learning environment.$$\n1. Career Counseling Module (Aptitude Test, Interest Test, Career Guidance)\r\n2. Degree & Career Scope (Job Market, Future Trends, Employment Opportunities)\r\n3. University & Degree Recommendation System (Personalized Suggestions, Smart Matching, Career Alignment)\r\n4. Comparison Tool (Side-by-Side Analysis, Universities, Degrees, Rankings)\r\n5. Student Community & Discussion Forum (Peer Interaction, Knowledge Sharing, Guidance Exchange)$$\n1. Student Profile & Dashboard (Profile, Progress Tracking, Personalized Dashboard)\r\n2. University Information Module (Programs, Fee Structure, Merit Lists, Facilities)\r\n3. AI Chatbot Career Counselor (AI Counseling, Aptitude Test, Recommendations)\r\n4. Notification Alerts (Reminders, Deadlines, Updates)\r\n5. Admissions & Testing Module (Entry Tests, Admission Process, Results)$$\n$$\n1. Career Guide - Hybrid App for students to select their Career in University\r\n2. Student Career Consultant$$\nAI-powered Career Counseling: Automated aptitude\/interest test with personalized\r\ndegree and career recommendations (instead of only manual counselor guidance).$$\nDegree & Career Scope: Provides insights into job market demand, industry requirements, and future career trends for informed decision-making.$$\nStudent Community & Discussion Forum: A peer-to-peer platform where students can discuss admissions, accommodation, test prep, and get guidance from current university students.","comments":" $$ Students will used scarping for data collection and also some APIs. As modern webistes usually don\u2019t allow scraping, its students responsibility to get realtime data, and perform analysis and decsions on the acquired dataset.","isDraft":0,"status":2,"created_at":"2025-10-06 17:21:41","updated_at":"2025-10-15 15:27:47","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":978,"project_id":1355,"title":"Drive Doctor: Smart Vehicle Maintenance Assistant","prob":"Car owners frequently encounter difficulties when their vehicles develop problems. While some issues are minor and can be resolved without professional help, most owners lack the technical knowledge to identify and fix them. As a result, they rely heavily on mechanics.\r\nTraditional OBD (On-Board Diagnostics) applications provide limited assistance, as they only display Diagnostic Trouble Codes (DTCs) and basic descriptions. These codes are often too technical for average users to understand, making it difficult to pinpoint the actual faulty part. This creates a gap between the data provided by the vehicle\u2019s ECU and the practical knowledge required by the vehicle owner.\r\nTherefore, there is a strong need for an intelligent and user-friendly diagnostic system that not only decodes errors but also identifies the faulty components, visualizes the problem clearly, and provides solutions that empower vehicle owners to handle minor issues themselves while assisting mechanics with accurate diagnostics for complex repairs.","description":"Drive Doctor \u2013 An Intelligent Vehicle Diagnostic Application is a mobile solution developed to simplify vehicle diagnostics by making technical information accessible to everyday car owners. Modern vehicles rely on Electronic Control Units (ECUs), which generate Diagnostic Trouble Codes (DTCs) when faults occur. However, traditional OBD-II scanners and applications typically display these codes in highly technical terms, making them difficult for non-technical users to interpret. This project addresses that gap by providing an intelligent system that not only decodes error codes but also enhances understanding through visualizations and guided repair instructions.\r\nThe application connects to vehicle ECUs via OBD-II adapters, retrieves diagnostic data, and translates raw error codes into clear, user-friendly descriptions. It further identifies the exact faulty components and employs interactive 3D simulations to visually highlight the affected parts. Alongside this, it offers step-by-step guidance for resolving minor issues independently and generates detailed diagnostic reports to assist mechanics with more complex problems. This combination reduces unnecessary visits to workshops, saving both time and money, while also strengthening communication between car owners and professionals.\r\nBy making vehicle diagnostics more transparent and accessible, Drive Doctor empowers users to take greater control of their vehicle maintenance, supports efficient and accurate repairs, and encourages preventative care. Ultimately, the system contributes to improved safety, reliability, and overall efficiency within the automotive industry.\r\nThis project is significant for multiple stakeholders:\r\n\u2022\tFor Vehicle Owners:\r\no\tEmpowers them to understand their vehicles better.\r\no\tReduces unnecessary expenses on minor issues.\r\no\tSaves time by solving small problems independently.\r\n\u2022\tFor Mechanics:\r\no\tProvides accurate diagnostic reports, making troubleshooting faster.\r\no\tMinimizes guesswork and ensures precise repairs.\r\n\u2022\tFor the Automotive Industry:\r\no\tPromotes smarter use of OBD-II technology.\r\no\tEncourages preventative maintenance, reducing breakdown risks.\r\no\tContributes to improved safety and vehicle reliability.$$\n1.\tConnectivity \r\nThis module will handle communication between the OBD scanner and the app via Wi-Fi or Bluetooth. Also, this module will be responsible for session management and pairing. This module also supports ELM327(a connectivity chip in OBD II)\r\n2.\tVehicle Data retrieval \r\nThis module handles data retrieval from the vehicle in the form of standard OBD parameter IDs, such as Speed, RPM, Fuel level, Throttle position, Coolant temperature, Intake temperature, etc. Data is retrieved in the form of hex codes, which are further translated into a human-readable format. This module also supports custom parameter IDs for different vehicles.\r\n3.\tDiagnostics\r\nThis module retrieves the DTCs from the ECU logs and translates the codes into a human-readable format. Also, clear system logs after servicing.\r\n4.\tReal-time monitoring and visualization\r\nThis module records the vehicle parameter IDs such as Speed, RPM, Fuel level, Throttle position, Coolant temperature, Intake temperature, etc. Logging & reporting \r\n5.\tMaintenance alerts\r\nThis module records the maintenance schedules for the vehicle and notifies the user of future maintenance. \r\n6.\tUser & vehicle Profile management and security\r\nThis module handles the most integral part of the application, which manages user and vehicle profile and performs further operations.\r\n7.\tDriving Behavior Analysis\r\nThis module will analyze the driving behaviors of drivers based on harsh braking and speeding using the built-in parameters like speed and PRM fluctuations and alert the user using alarms and alerts.\r\n8.\tAR camera integration \r\nThis module will use the user\u2019s device camera and then direct the user to the faulty part, and then suggest a solution to fix that part. This module will also implement an AR-based user guide to guide non-technical and new users about the system workflow.\r\n9.\t3D Animated Repair Guide\r\nThis module will implement the 3D-based animated solutions against the retrieved errors and guide to user to easy solutions that do not need professional assistance (for easily resolvable problems). This module will also implement demo version of system which contain the instructions to use the application.$$\nIn Drive Doctor, I shall develop the modules which include:\r\no\tOBD Connectivity module\r\no\tVehicle data retrieval module \r\no\tReal-time monitoring and visualization module.\r\no\tDriving Behavior Analysis\r\no\t3D Animated Repair Guide$$\nIn Drive Doctor I shall develop the modules which include:\r\no\tDiagnostics module\r\no\tMaintenance alerts\r\no\tUser authentication and vehicle profile management & security\r\no\tAR Camera integration.$$\n$$\n1. LubriCare$$\n1.\tDTC Translation$$\n2.\tSmart Repair Visualization$$\n3.\tAugmented Repair Assistance","comments":" $$ Student have added 3 modules, AI based driving behavior analysis, 3d animation based repairing guide and AR camera integration to identify the error. Please double check the feaibility of AR based fault detection. $$ Student have added 3 modules, AI based driving behavior analysis, 3d animation based repairing guide and AR camera integration to identify the error. Please double check the feaibility of AR based fault detection.","isDraft":0,"status":2,"created_at":"2025-10-04 11:13:34","updated_at":"2025-10-17 08:50:22","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":959,"project_id":1357,"title":"EvaluTech \u2013 AI-Powered Interview & Hiring Management Platform for \u201cFactory Web Services\u201d","prob":"Factory Web Services (FWS), a growing organization in Pakistan, struggles with large applicant volumes, biased evaluations, and hiring delays. Manual recruitment consumes HR time and causes poor candidate-job matching. EvaluTech solves this by offering an AI-powered hiring platform that automates resume screening, conducts automated live interviews between Employer and Candidate (with panel) directly on the platform, generates transcripts, and provides analytics. Applicants can create profiles, submit resumes even without active job postings, and later receive recommendations through semantic skills matching. With customizable pipelines complete assessment evidence, and proctoring features, EvaluTech streamlines hiring, ensures fairness, and helps FWS recruit faster while reducing overhead.","description":"EvaluTech is an AI-powered hiring platform developed for \u201cFactory Web Services\u201d based on their specific requirements. It streamlines the recruitment process, making it faster, smarter, and more efficient \u2014 not only for FWS but also for other organizations seeking a modern hiring solution. Employers can create job posts and set custom hiring steps such as resume screening, pre-screening questionnaires, aptitude\/MCQ tests, technical assessment uploads, or interviews. Candidates can build profiles, apply for jobs, and complete interviews directly on the platform. Even without an active job, candidates can submit resumes, and a recommendation engine will later notify them about relevant openings. The system supports live employer\u2013candidate interviews with AI-generated transcripts and analytics. EvaluTech can also generate job descriptions, interview questions, and feedback using AI, but all AI content is editable by HR\u2014 nothing proceeds without employer approval. Importantly, it uses semantic-based skills matching, ensuring profiles are matched to job descriptions by understanding meaning (not just keywords), improving fairness and accuracy.\r\nTo support decision-making, EvaluTech offers a Complete Assessment Evidence Tab for each candidate \u2014 showing profile details, skills match, test scores, and AI insights from resume analysis such as strengths, experience evaluation, and final recommendations. Employers can also review MCQ results, interview transcripts, ensuring decisions are backed by clear evidence.\r\nUnlike existing systems like \u201cAUTOHIRE.WORKS\u201d, only a non-operational landing page with login\/signup(non-operational), buttons not functional, can't even signup in system.\r\n\r\nand \u201cAUTOHIRE.CV\u201d, limited to employer-side features with manual resume uploads, minimal AI, no candidate portal, no customizable pipelines, and its interview service is inactive.\r\n\r\nand \u201cVANHACK\u201d, focuses only on global tech hiring, offers candidate matching but lacks live interviews, customizable pipelines, and proctoring. \r\n\r\nOur EvaluTech delivers a complete end-to-end solution \u2014 covering candidate and employer sides, AI-automation to reduce manual work, customizable hiring pipelines, live interviews with transcripts.$$\ni.\tCANDIDATE PROFILE & APPLICATION PORTAL \u2013 Profile creation, resume upload, job browsing and applications, job recommendations, application tracking.\r\nii.\tEMPLOYER DASHBOARD & PIPELINE MANAGER \u2013 Job posting, applicant management, customizable hiring pipelines, AI-generated job descriptions and questions, candidate history, interview scheduling, centralized tracking. (Employers can always edit AI-generated job descriptions or create them manually \u2014 no AI-generated content is finalized without employer\/HR approval.)\r\niii.\tLIVE INTERVIEW & TRANSCRIPT SYSTEM \u2013 Employer\u2013candidate live interviews with transcript generation and storage, providing with a record of what was discussed in the interview as well as the strengths and weaknesses of candidate.\r\niv.\tCANDIDATE EVALUATION & REPORTING MODULE \u2013 Candidate ranking, semantic AI-based job matching, evaluation reports, scoring metrics, CSV\/PDF export, complete assessment evidence tab.\r\nv.\tPROCTORING & ACTIVITY MONITORING MODULE \u2013 Tab switching and inactivity detection, session integrity validation, proctoring reports, activity logging during online tests.\r\nvi.\tNOTIFICATIONS, STATUS TRACKING & HR RECORD MANAGEMENT \u2013 Real-time job\/application notifications, application status updates, interview scheduling alerts multi-channel delivery, reminders, and centralized applicant records\/history for HR with CSV\/PDF export.$$\n(i)\tCandidate Profile & Application Portal \u2013 Ahad will build the portal where candidates create professional profiles, upload resumes, browse jobs, and track applications, and apply to multiple jobs with ease and accuracy.\r\n(ii)\tProctoring & Activity Monitoring Module \u2013 He will ensure assessment integrity by developing proctoring features that monitor candidate activity, prevent cheating, and provide employers with fair evaluation insights.\r\n(iii)\tNotifications & Applicant Record Management \u2013 Ahad will build the notification system to keep candidates and employers updated in real time while allowing recruiters to maintain structured records and track applications efficiently.$$\n(i)\tEmployer Dashboard & Pipeline Manager \u2013 Raheel will create a dashboard where employers can post jobs, track candidates, schedule interviews, and customize hiring pipelines to suit specific recruitment needs.\r\n(ii)\tLive Interview & Transcript System \u2013 He will implement automated live interviews with employer, integrating transcript generation and provide analytics.\r\n(iii)\tCandidate Evaluation & Reporting Module \u2013 Raheel will develop analytics-driven evaluation features, including semantic job matching, to provide employers with candidate performance reports, scoring metrics, and actionable insights that support evidence-based hiring decisions.$$\n$$\nSmart Hire$$\n1. AI-Powered Live Interviews (Employer-Candidate) with Transcript Generation \u2013\r\nAlso, unlike Smart Hire, this is an industry project with Factory Web Services so it is practical and company-driven.$$\n2. Generative AI Automation \u2013 Auto-creation of job descriptions, interview questions, quizzes, and candidate feedback.$$\n3. Customizable Hiring Pipelines \u2013 Employers can set job stages: CV screening, pre-screening questionnaire, aptitude\/MCQ test, technical assessment upload, and live interview with transcript.","comments":" $$ Fair amount of the modules are added to system. It is recommended to vivusllay present the candidate's skills relevance to the ad. Or even sort the CVs w.r.t the relevance of the job.","isDraft":0,"status":2,"created_at":"2025-09-29 15:40:42","updated_at":"2025-10-15 15:24:31","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1004,"project_id":1360,"title":"HealthSync","prob":"Currently, patients of Atta Pharmacy must visit physically to purchase medicines, which limits healthcare accessibility. To improve services, Atta Pharmacy aims to expand into a clinic by combining pharmacy, medical consultation, and digital healthcare under one platform. Patients will be able to book appointments online, meet doctors, and receive prescriptions with required medicines in a single visit. Those unable to visit can order medicines online with home delivery. Doctors will have access to patient history, prescriptions, and lab reports, ensuring better care and accurate treatment. To enhance scalability and usability, the system will also include a Mobile App module for patients and online ordering, designed as a generic product supporting multiple virtual stores with a Super Admin to manage all clinics and stores.","description":"HealthSync is a web-based and mobile-enabled system for Atta Pharmacy that unifies pharmacy, clinic, lab, and delivery services on a single platform. Patients can register, manage profiles, order medicines, and book doctor appointments online. The Patient and Online Ordering Modules will be available via a Mobile App, designed as a generic, multi-clinic platform where virtual stores can be added and managed by a Super Admin. Doctors and patients' profiles will be created by the admin to ensure accurate records and proper onboarding. Patients can view prescriptions, purchase history, and lab reports, while receiving timely notifications for appointments, refills, and deliveries. Doctors can manage schedules, upload prescriptions, and recommend lab tests. The admin oversees medicines, procurement, analytics, and feedback. The system also includes procurement workflows and a sticker-based medicine marking system to ensure authenticity. Security features such as role-based access, encryption, and logging make HealthSync a complete, scalable healthcare solution.$$\nIn HealthSync: Online Pharmacy & Clinic Management System, Industrial Project for Atta Pharmacy, the modules are: (1) Patient Module (Mobile App) Personal space for patients to sign up, log in, browse\/order medicines, and book appointments. Accessible primarily via Mobile App for better usability. (2) Doctor Module: Digital assistant for doctors to manage schedules, view patients' history, and upload prescriptions. (3) Admin Module (Super Admin & Multi-Clinic Management): Control center to manage medicines, doctors, profiles, procurement, and reports. Admin will also add doctors' and patients' profiles. (4) Delivery Module: Logistics partner to approve orders, assign delivery staff, and send updates. (5) Lab Module: Upload and manage lab reports. (6) Smart Prescription & Refill Module: Tracks prescription expiry, alerts for refills. (7) Notification & Communication Module: Keeps patients informed of appointments, orders, and reminders. (8) Security & Authentication Module: Provides secure login, role-based access, encryption, and activity logging. (9) Procurement & Medicine Marking Module: Ensures stock management, sale and procurement workflows, and medicine sticker labeling for authenticity.$$\nIn HealthSync: I shall develop the modules which include: (i) Patient Module, (ii) Lab Module, (iii) Smart Prescription & Refill module, (iv) Security & Authentication Module.$$\nIn HealthSync: I shall develop the modules which include: (i) Doctor Module, (ii) Admin Module, (iii) Delivery module, (iv) Notification & Communication Module, (v) Procurement & Medicine Marking Module.$$\n$$\n(i)\tOnline Medicine Store (Medicine to Home)\r\n(ii)\tAndroid based Online Medicine ordering$$\n1. Marketplace with multiple clinics & pharmacies under Super Admin$$\n2. Mobile App for Patient & Ordering modules$$\n3. Prescription Refill Reminders (Improves Patient care)","comments":" $$ The revised HealthSync proposal effectively addresses the committee\u2019s earlier feedback by transforming a limited pharmacy management concept into a comprehensive, product-oriented healthcare platform. The inclusion of a mobile app for patients, super admin for multi-clinic management, and a procurement and medicine marking module demonstrates clear improvement in scope, scalability, and innovation. The system now presents a cohesive integration of pharmacy, clinic, and lab functionalities, offering genuine value for healthcare digitization. While the overall revision is strong, the technical feasibility of procurement workflows and sticker-based verification should be elaborated further to ensure practical implementation. Overall, the project is now substantially improved and can be approved with minor clarifications.","isDraft":0,"status":2,"created_at":"2025-10-08 14:40:57","updated_at":"2025-10-17 08:53:57","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1050,"project_id":1427,"title":"PMNH BioConnect","prob":"Pakistan has a rich biodiversity consisting of thousands of plant, animal, and mineral species. However, most data exists in physical records and scattered research papers that are difficult to access. Researchers and students struggle to find, compare, and analyze biodiversity information efficiently.\r\nThere is no centralized digital platform or mobile application that provides both species information and automatic AI-based identification. This project solves that problem by developing a cloud-based biodiversity database and mobile application integrated with an AI classifier trained on PMNH datasets.","description":"This project focuses on designing, developing, and deploying a mobile application integrated with a cloud-based biodiversity database for managing Pakistan\u2019s natural species data. The system uses Firebase Firestore as the primary NoSQL cloud database to store structured information such as species names, classification data, descriptions, GPS coordinates, and associated metadata.\r\nMultimedia assets\u2014such as images, videos, and audio recordings\u2014are stored in Firebase Storage, which provides scalable, secure, and cost efficient cloud storage. User authentication and role management (Admin, Researcher, Visitor) are handled through Firebase Authentication.\r\nThe mobile application, developed in Flutter, communicates directly with Firebase services for real-time operations including browsing species data, uploading media, retrieving historical information, and synchronizing offline data.\r\nA key module of the system is the AI-based Image Analysis & Classifier, which uses a custom-trained deep learning model to identify species from user-uploaded images. The model is hosted either locally inside the app or via Firebase Cloud Functions, enabling real-time identification. The classifier initially supports 10 species from each biological kingdom and can be expanded later.\r\nThis entire Firebase-based architecture ensures scalability, industry-level performance, lower operational costs, and ease of deployment for public use by researchers, visitors, and conservation teams.$$\n1. User Authentication Module\r\nHandles secure login, registration, and role-based access using Firebase Authentication. Supports email\/password, Google login, and token-based access.\r\n2. Data Entry & Management Module\r\nAdmins and researchers manage biodiversity data stored in Firestore. Includes adding, editing, and deleting species information with associated metadata.\r\n3. Search & Retrieval Module\r\nOptimized Firestore queries allow users to search species by taxonomy, names, or GPS regions with fast real-time results.\r\n4. Category Management Module\r\nMaintains the biological hierarchy (kingdom \u2192 species) using structured Firestore collections and subcollections.\r\n5. Mobile Application Module\r\nFlutter-based app providing:\r\n\u2022\tOffline support\r\n\u2022\tCamera \/ gallery uploads\r\n\u2022\tGPS tagging\r\n\u2022\tReal-time synchronization via Firebase SDKs\r\n\u2022\tData caching\r\n6. Image Analysis & Classifier Module\r\nUses a trained AI model to classify species from uploaded images. Implemented via:\r\n\u2022\tFirebase Cloud Functions (server-side inference) or\r\n\u2022\tOn-device lightweight model using TensorFlow Lite.\r\nThe classifier returns species name, habitat, and confidence score.\r\n7. Reporting & Analytics Module\r\nFirestore data is analyzed to generate biodiversity insights such as species distribution, frequently recorded specimens, and region-wise diversity. Firebase Functions can generate periodic summaries.\r\n8. Backup & Recovery Module (Updated)\r\nSince Firebase is cloud-based, automated backups include:\r\n\u2022\tFirestore export backups\r\n\u2022\tFirebase Storage backup snapshots\r\n\u2022\tOptional Google Cloud scheduled backups$$\nResponsible for designing the Firestore database structure, managing Firebase Storage media workflow, implementing Cloud Functions for API logic, and integrating the AI model with Firebase backend services. Also responsible for analytics and reporting.$$\nResponsible for developing the Flutter mobile application, connecting it with Firebase services, implementing media upload, GPS tagging, species recognition interface, and ensuring smooth UI\/UX, performance, and offline access for users.$$\n$$\nNull$$\n$$\n$$\n","comments":" $$ tudent have included the image identification and tagging module to automate the process, as suggested. This automation tagging will be applied to at least 10 species from each biological hierarchy along with the manual tagging","isDraft":0,"status":2,"created_at":"2025-11-19 12:44:26","updated_at":"2025-11-21 11:41:35","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":985,"project_id":1363,"title":"Zarkhez (Automated Irrigation Control)","prob":"In Pakistan, farmers face serious challenges like water wastage, high electricity costs, motor failures, and inefficient irrigation scheduling. Most tube-wells run manually without accurate soil or weather data, leading to over-irrigation, crop damage, and resource loss. Shared tube-wells also create disputes due to lack of transparent billing. This FYP solves these issues by providing an IoT and AI-powered smart irrigation and energy management system that ensures precision watering, automatic motor safety, predictive maintenance, and fair multi-user billing. It directly addresses low farmer literacy, high operational costs, and unsustainable water usage, making farming efficient, reliable, and affordable.","description":"This project presents an IoT and AI-based Smart Irrigation & Energy Management System designed for precision agriculture in Pakistan. The system integrates ESP32 microcontrollers with MQTT\/HTTP protocols for real-time motor control, telemetry, and operational safety. An AI-driven scheduling engine dynamically adjusts irrigation using soil moisture sensors, weather APIs, and rainfall forecasts, ensuring optimized water usage. Machine learning models predict irrigation efficiency and water wastage, promoting sustainable farming. Energy monitoring modules track voltage and current consumption, enabling multi-user billing with automated tariff calculations. Advanced motor protection circuits safeguard against voltage fluctuations and overload. Predictive maintenance logs runtime hours to prevent failures. A bilingual farmer mobile app offers remote control, billing, and schedules, while the admin console manages multiple clients. Transparent audit logs ensure trust, accountability, and scalability.$$\n1. Smart Motor Control & IoT Layer\r\n\u2022\tRemote ON\/OFF via ESP32 microcontroller.\r\n\u2022\tMQTT\/HTTP protocols for real-time communication.\r\n\u2022\tLive motor status visible in farmer mobile app.\r\n2. Autonomous Irrigation Scheduling Engine\r\n\u2022\tFarmers can easily set automatic irrigation schedules through a user-friendly interface.\r\n\u2022\tThe system automatically executes motor start\/stop operations as per the defined schedule.\r\n\u2022\tThis ensures irrigation is always on time, reducing delays.\r\n3. Smart Soil (Soil Sensor), Crop Intelligence & Water Forecasting\r\n\u2022\tReal-time deep soil sensor monitoring.\r\n\u2022\tKnowledge-based engine for crop-specific water need calculation.\r\n\u2022\tWeather API and local dataset integration for adaptive recommendations.\r\n\u2022\tML-based predictions on irrigation efficiency and water wastage.\r\n\u2022\tHistorical and real-time log analysis.\r\n\u2022\tData-driven insights for sustainable farming.\r\n4. Smart Energy, Billing & Event Logging Subsystem\r\n\u2022\tReal-time monitoring of voltage and current usage.\r\n\u2022\tMulti-user billing support (shared tube-well scenarios).\r\n\u2022\tComprehensive time-stamped logs of irrigation and motor operations.\r\n\u2022\tTransparent audit trail accessible by farmers anytime.\r\n\u2022\tEnhances trust, accountability, and decision-making.\r\n5. Motor Protection, Safety & Predictive Maintenance\r\n\u2022\tShut-down on unsafe voltage (180V \/ 240V).\r\n\u2022\tCurrent overload detection and protection (>12A).\r\n\u2022\tAuto Mode: fully self-protective.\r\n\u2022\tManual Mode: alerts to farmer before shutdown.\r\n\u2022\tRuntime logging of motor operational hours.\r\n\u2022\tService reminders after defined thresholds (e.g., 500 hours).\r\n\u2022\tPrevents unexpected breakdowns and increases motor life.\r\n6. Super Admin & Client Management Console\r\n\u2022\tCentralized dashboard for service providers.\r\n\u2022\tDevice provisioning and remote configurations.\r\n\u2022\tBilling plan creation and management.\r\n\u2022\tBusiness-level scalability with analytics.\r\n7. Farmer Mobile Application (Bilingual)\r\n\u2022\tDual language support: Urdu + English.\r\n\u2022\tFeatures: motor control, scheduling, billing, crop input, notifications.$$\n\u2022\tSmart Motor Control & IoT Layer \r\n\u2022\tAutonomous Irrigation Scheduling Engine\r\n\u2022\tSmart Soil & Crop Intelligence Layer \r\n\u2022\tAI-Powered Water Efficiency & Forecasting\r\n\u2022\tMotor Protection & Safety Management$$\n\u2022\tSmart Energy & Billing Subsystem\r\n\u2022\tPredictive Maintenance & Alerts \r\n\u2022\tSuper Admin & Client Management Console \r\n\u2022\tComprehensive Event Logging & Audit Trail \r\n\u2022\tUrdu Language Support$$\n$$\n$$\n$$\n$$\n","comments":" $$ Enough modules are added and multi source driven decsion is proposed which is good. The proposed modules must be developed with their corresponding usecase.","isDraft":0,"status":2,"created_at":"2025-10-07 11:21:56","updated_at":"2025-10-15 15:25:42","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1008,"project_id":1361,"title":"Skill Swap","prob":"* \r\nMany people want to learn new skills but cannot afford paid courses, while others have skills they could teach but lack a platform to share and exchange them. Existing learning platforms are either costly or one-directional, where only instructors teach and learners pay. SkillSwap solves this by enabling people to exchange skills directly with one another \u2014 e.g., \u201cteach me video editing, I\u2019ll teach you web design.\u201d It eliminates financial barriers, promotes knowledge sharing, and builds a global network of learners and teachers.","description":"SkillSwap is a web-based platform that allows people to exchange skills directly with one another. Users can create profiles showcasing the skills they can teach and the skills they want to learn. The system uses an AI-based matching algorithm to connect users with suitable partners for mutual learning. After matching, they can chat, schedule sessions, and track progress. Users earn credits for teaching others, which can be used to learn new skills in return. The platform also includes reviews, ratings, and admin monitoring to ensure a safe and high-quality learning environment. In short, SkillSwap promotes free, fair, and skill-based learning without relying on traditional paid courses.$$\n1.\tUser Management Module:\r\nHandles user registration, login, profile creation, and skill listings. Users can specify which skills they can teach and which ones they want to learn. It also manages authentication, authorization, and profile updates.\r\n\t2.\tSkill Matching Module:\r\nUses AI-based algorithms to match users with others who have complementary skills. It suggests the most suitable partners for learning or teaching based on user profiles, interests, and skill categories.\r\n\t3.\tCredit & Reward Module:\r\nIntroduces a credit-based system where users earn credits by teaching others and spend them to learn new skills. It ensures fair exchange and motivates continuous participation.\r\n\t4.\tChat & Scheduling Module:\r\nProvides real-time messaging and session scheduling between matched users. It helps them communicate, set timings for lessons, and coordinate learning activities easily.\r\n\t5.\tReview & Rating Module:\r\nAllows users to give feedback after each learning or teaching session. Ratings help maintain quality, build trust, and improve future matching accuracy.\r\n\t6.\tAdmin & Analytics Module:\r\nEnables administrators to manage users, monitor activities, and analyze platform performance. It includes dashboards for insights such as popular skills, active users, and learning trends.$$\n1.\tUser Management Module:\r\nHandles user registration, login, profile creation, and skill listings. Users can specify which skills they can teach and which ones they want to learn. It also manages authentication, authorization, and profile updates.\r\n\t2.\tSkill Matching Module:\r\nUses AI-based algorithms to match users with others who have complementary skills. It suggests the most suitable partners for learning or teaching based on user profiles, interests, and skill categories.\r\n\t3.\tCredit & Reward Module:\r\nIntroduces a credit-based system where users earn credits by teaching others and spend them to learn new skills. It ensures fair exchange and motivates continuous participation.$$\n4.\tChat & Scheduling Module:\r\nProvides real-time messaging and session scheduling between matched users. It helps them communicate, set timings for lessons, and coordinate learning activities easily.\r\n\t5.\tReview & Rating Module:\r\nAllows users to give feedback after each learning or teaching session. Ratings help maintain quality, build trust, and improve future matching accuracy.\r\n\t6.\tAdmin & Analytics Module:\r\nEnables administrators to manage users, monitor activities, and analyze platform performance. It includes dashboards for insights such as popular skills, active users, and learning trends.$$\n$$\n$$\n$$\n$$\n","comments":" $$ The idea is good and students are required to carefully plan the flow of the project. For Example how this palteform can be attractive for swapping the skills and API based functanlity must be provided like online meeting session, chat and etc $$ The idea is good and students are required to carefully plan the flow of the project. For Example how this palteform can be attractive for swapping the skills and API based functanlity must be provided like online meeting session, chat and etc","isDraft":0,"status":2,"created_at":"2025-10-08 16:13:45","updated_at":"2025-10-17 08:51:00","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":982,"project_id":1387,"title":"Share4Good- A Smart Platform for Donors and Recipients","prob":"In our society, many needy individuals struggle to access necessities such as clothing, electronics, education, financial assistance and other essentials. On the other hand, many people are willing to donate but there is lack of transparent, trustworthy, and organized system to ensure their donations reach genuine recipients. Existing donation systems are often manual, NGO-based, or lack transparency, which creates doubts among donors and difficulties for recipients.\r\nShare4Good aims to bridge this gap by providing a digital platform where donors and recipients can connect transparently. The system ensures proper verification, secure transactions, and fair distribution. It also introduces an auction mechanism for high-quality or unique donated products, where funds generated are further used for verified charitable purposes.","description":"Project Description:\r\nShare4Good is a web-based platform that allows users to participate both as donors and recipients. The system is managed and controlled by an admin, who verifies all activities for transparency.\r\nDonors: \r\n\u2022\tRegister on the platform\r\n\u2022\tDonate products or cash\r\n\u2022\tProvide details of their donation i.e.: to define purpose of their donation\r\n\u2022\tUpload Images of their donations.\r\n\u2022\tThe needs mentioned by needy people will be shown to the donor panel.\r\nRecipients :\r\n\u2022\tRegister by submitting personal details and required verification documents (such as salary slips, ID cards, or supporting documents).\r\n\u2022\tAfter verification, they can request products or financial help(e.g., education support, wedding expenses, medical help).\r\nAdmin:\r\n\u2022\t verify donors and recipients,\r\n\u2022\t approve donations, \r\n\u2022\tmanage distribution, and oversee all products, cash, and auction activities.\r\nThe platform operates in three modules: Product Module, Cash Module, and Auction Module.$$\n1.\t Product Module\r\n\u2022\tDonors register and upload details\/images of products they want to donate.\r\n\u2022\tAdmin verifies and approves the donated product. All donated products are collected under admin\u2019s control.\r\n\u2022\tProducts will be categorized into their appropriate category.\r\n\u2022\tRecipients register and upload verification documents (salary slip, ID, etc.). Once approved, he\/she can search for available products.\r\n\u2022\tIf a product is available, the recipient can send a request. If not, the system shows \u201cNot Available.\u201d\r\n\u2022\tRecipient requests go through the admin, and once approved, the product is delivered to the recipient.\r\n2.\tCash Module\r\n\u2022\tDonor specifies the amount of cash donation along with the purpose (e.g., education support, wedding expenses, medical help).\r\n\u2022\tAdmin verifies and collects the cash. All donations are safely stored with their declared purposes.\r\n\u2022\tRecipient register, provides documents, and applies for cash support, also stating their reason\/purpose.\r\n\u2022\tIf matching funds are available, admin approves the request and transfers the cash to the recipient via bank account or other methods.\r\n\u2022\tIf funds are not available, the system displays \u201cNot Available.\u201d\r\n3.\tAuction Module\r\n\u2022\tDonors donate products as usual.\r\n\u2022\tAdmin evaluates donated items. If a product is of high quality or unique, the admin shifts it into the Auction Category.\r\n\u2022\tAdmin sets a specific time limit for bidding.\r\n\u2022\tUsers place bids on the product during the auction window.\r\n\u2022\tAt the end of the auction, the highest bidder wins, and the product is delivered to them by admin.\r\n\u2022\tThe funds collected from auctions are stored and utilized for fundraising or charitable purposes.$$\n1) Product Module:\r\n\u2022\tDonors register and upload details\/images of products they want to donate.\r\n\u2022\tAdmin verifies and approves the donated product. All donated products are collected under admin\u2019s control.\r\n\u2022\tProducts will be categorized into their appropriate category.\r\n\u2022\tRecipients register and upload verification documents (salary slip, ID, etc.). Once approved, he\/she can search for available products.\r\n\u2022\tIf a product is available, the recipient can send a request. If not, the system shows \u201cNot Available.\u201d\r\n\u2022\tRecipient requests go through the admin, and once approved, the product is delivered to the recipient.\r\n2) Auction Module$$\n1) Cash Module:\r\n\u2022\tDonor specifies the amount of cash donation along with the purpose (e.g., education support, wedding expenses, medical help).\r\n\u2022\tAdmin verifies and collects the cash. All donations are safely stored with their declared purposes.\r\n\u2022\tRecipient register, provides documents, and applies for cash support, also stating their reason\/purpose.\r\n\u2022\tIf matching funds are available, admin approves the request and transfers the cash to the recipient via bank account or other methods.\r\n\u2022\tIf funds are not available, the system displays \u201cNot Available.\u201d\r\n2) Auction Module$$\n$$\nGIVEHOPE (Helping Poor People by Donation)$$\n1) Auction-based Donation System\r\n\u2022\t Donated products can be auctioned, recipients place bids, and proceeds go to verified charity causes.$$\n2) Dual Donation Handling (Cash + Product)\r\n\u2022\tThe system manages both product and cash donations separately, offering more flexibility for donors and recipients.$$\n3.\tAI-powered Illegal Product Detection\r\n\u2022\tAn AI model automatically detects and blocks illegal items uploaded by donors.","comments":" $$ The team has successfully addressed all major concerns raised by the committee. They expanded the project scope beyond a simple bidding system, clearly separated the auction, cash, and product donation modules, and explained AI-based illegal product detection feasibility. The system structure now ensures transparency, verification, and a more meaningful social impact.","isDraft":0,"status":2,"created_at":"2025-10-06 13:35:08","updated_at":"2025-10-17 11:37:39","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1024,"project_id":1377,"title":"Road Damage Detection System","prob":"Road safety and maintenance are major challenges as damaged roads lead to accidents, traffic delays, and costly repairs. Traditional inspection is manual, time-consuming, and prone to human error. This project solves the problem by providing an AI-powered automated road damage detection system. Using computer vision (YOLOv8) and deep learning, the application can identify and classify multiple types of road damages such as potholes, cracks, and damaged markings in real-time from images, videos, or live camera feeds. It also generates PDF reports with damage counts and their exact GPS coordinates, helping authorities locate and assess road damages accurately. This system reduces manpower, saves time, and ensures accurate, data-driven decisions for better infrastructure maintenance. Every video is split into multiple frames at a fixed rate. The YOLOv8 model runs inference on each frame independently to detect damages.","description":"The Road Damage Detection Application is an AI-based solution that uses YOLOv8 object detection integrated with Stream lit to detect and classify different types of road damages in real-time. The system supports image upload, video upload, and live camera feed for detection. It can identify seven types of road damages: Alligator Cracks, Damaged Crosswalk, Damaged Paint, Longitudinal Cracks, Manhole Cover, Potholes, and Transverse Cracks. After detection, the application generates a PDF report summarizing the total damages and type-wise counts. The system automates road inspection to reduce effort, improve accuracy, and deliver quick insights for maintenance. The project follows these steps:\r\n1.\tDataset Collection & Preprocessing \u2013 Road damage dataset from Kaggle and local images will be used.\r\n2.\tModel Training \u2013 YOLOv8 model trained on the dataset for accurate classification.\r\n3.\tStream lit Web Interface \u2013 For users to upload images\/videos or access live camera feed.\r\n4.\tReal-Time Detection & Tracking \u2013 Damage detection on frames using YOLOv8 + SORT tracking.\r\n5.\tReport Generation \u2013 Automated PDF with detection details and their exact GPS coordinates\r\n6.\tDeployment \u2013 Stream lit app deployed locally or on cloud for accessibility.$$\n1.\tDataset & Preprocessing Module \u2013 Collect, clean, and augment road damage images.\r\n2.\tModel Training Module \u2013 Train YOLOv8 on 7 damage categories.\r\n3.\tImage Inference Module \u2013 Upload images and detect damages.\r\n4.\tVideo Inference Module \u2013 Upload video for frame-by-frame detection and counting.\r\n5.\tLive Camera Module \u2013 Real-time detection using webcam feed.\r\n6.\tObject Tracking Module \u2013 Apply SORT algorithm to track damages across frames.\r\n7.\tReport Generation Module \u2013 Generate and export PDF summaries.\r\n8.\tLocation\/coordinates of damage- Coordinates are stored and mapped to show damage locations.\r\n9.\tWeb UI Module \u2013 Stream lit-based interface for users.\r\n10.\tDeployment & Testing Module \u2013 Deploy on local\/cloud and validate performance.$$\nIn RDS, I will develop a module comprising Dataset & Preprocessing, Model Training, Report Generation, and Deployment. The Dataset & Preprocessing phase will involve data collection, cleaning, and transformation. Model Training will focus on building and validating predictive models. Report Generation will automate insights and results. Deployment will integrate the model into a user-accessible environment for real-time use.$$\nIn RDS, I will develop modules focusing on image, video, and live stream inference to enable real-time data processing and analysis. Additionally, I will implement object tracking algorithms to monitor and identify objects across frames accurately. To enhance usability and accessibility, I will design a streamlined web and mobile application UI, ensuring efficient interaction, intuitive design, and responsive performance.Damage coordinates are stored to map and visualize affected locations. These modules aim to deliver a seamless, real-time experience for end-users across platforms.$$\n$$\n$$\n1. Detects 7 different types of road damages instead of only potholes.$$\n2. Provides multi-input support (image, video, and live camera).$$\n3. Generates automated PDF reports with damage summaries for authorities and their exact GPS coordinates.","comments":" $$ The revised proposal shows clear progress in addressing several key comments from the committee. The inclusion of GPS coordinates and a dedicated location\/coordinate module directly responds to earlier feedback about lacking spatial information. The updated workflow now describes frame-based analysis, where each video is split into frames for YOLOv8 inference \u2014 effectively covering the previously unclear image processing approach. Additionally, the mention of object tracking (SORT algorithm) and multi-input support (images, videos, live camera) strengthens the technical depth and aligns with modern approaches for road damage detection.","isDraft":0,"status":2,"created_at":"2025-10-13 13:13:15","updated_at":"2025-10-17 09:51:18","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1007,"project_id":1418,"title":"CUI-Wah Job Fair Company and Admin Portal","prob":"Facilitating the companies during university job fairs required some administrative challenges that need to be automated. Companies struggle with verifying attendance, managing interview queues, and shortlisting candidates.\r\nAdministrators face difficulties in confirming arrivals, assigning rooms, and collecting feedback.\r\nWithout a centralized system, inefficiencies such as unauthorized participation, mismanagement, and scheduling conflicts occur that reduce the effectiveness of recruitment drives and increase the workload for organizers. To improve goodwill, an automated system is required for smooth scheduling and managerial issues at the venue.","description":"CUI Wah Job Fair Company and Admin Portal form the core of the \r\njob fair, ensuring smooth coordination between recruiters, \r\nstudents, and event organizers. The Company Portal allows \r\nrecruiters to digitally manage participation\u2014from marking \r\nattendance via QR code and specifying required skills to viewing \r\nstudent profiles, sending interview requests, and managing \r\ninterview queues automatically to prevent scheduling conflicts. \r\nCompanies can also exchange notifications, submit feedback, and \r\nreceive participation certificates.\r\nThe Admin Portal provides complete event control, enabling \r\norganizers to generate QR codes, monitor attendance, send \r\nnotifications, allocate interview rooms, and analyze performance \r\ndata. It includes analytics dashboards for tracking interviews, \r\nfeedback, and hiring outcomes, along with semester-wise record \r\nmaintenance for future reference and sharing. Together, both \r\nmodules digitalize the job fair process, ensuring efficiency, \r\ntransparency, and organized management.$$\nCompany Module (Web App)\r\n 1. Sign Up \/ Login \u2013 Secure company registration and login access.\r\n 2. QR Code Attendance Marking \u2013 Mandatory QR scan for verifying physical presence.\r\n 3. Confirm Arrival \u2013 Confirms participation after QR verification.\r\n 4. Enter Skills Required \u2013 Companies specify desired roles and skill sets.\r\n 5. View Student Profiles\/CVs & Projects \u2013 Access detailed student portfolios.\r\n 6. Send\/Receive Interview Requests \u2013 Communicate interview invitations with students.\r\n 7. Accept\/Reject Requests \u2013 Manage interview invitations flexibly.\r\n 8. Manage Interview Queue \u2013 Auto-managed scheduling system that prevents overlapping interviews; if a time clash occurs, the system automatically adjusts interview slots based on company and student availability.\r\n 9. Notifications \u2013 Send and receive notifications from the admin or students for updates and \r\nannouncements.\r\n 10. Submit Feedback \u2013 Provide structured feedback for event improvement.\r\n 11. Get Participation Certificates \u2013 Automatically receive digital participation certificates post-event.\r\n 12. Logout \u2013 Securely end the session to protect company data.\r\n 13. Hosting and Deployment \u2013 The company module will be hosted and deployed on the university campus network to provide reliable access during the job fair.\r\n\r\n\r\n Admin Module (Web App)\r\n1. Sign Up \/ Login \u2013 Secure access for authorized administrators.\r\n 2. Monitor Company Attendance \u2013 Track arrival and participation of companies in real time.\r\n 3. View Company & Student Details \u2013 Centralized access to all registered participants.\r\n 4. Notifications \u2013 Send and receive important announcements or updates to\/from students and companies.\r\n 5. Auto-Allocate Rooms \u2013 Automatically assigns interview rooms to companies; manual adjustments can be made if needed.\r\n 6. View Analytics \u2013 Displays statistics on interviews, company participation, student CVs, and hiring results.\r\n 7. Analyze Feedback \u2013 Review structured feedback from companies for quality improvement.\r\n 8. View Interview Status \u2013 Monitor final hiring outcomes (Hired\/Rejected).\r\n 9. Maintain Semester-Wise Data \u2013 Stores records of each semester\u2019s job fair, including student and company participation, hiring outcomes, and key statistics. Can be extended to include alumni data for future use in meetups, networking, and tracking career progress. Data can be shared with authorized personnel.\r\n 10. Logout \u2013 Ensures secure exit from the system to maintain confidentiality.\r\n 11. Hosting and Deployment \u2013 The admin module will be hosted and deployed on the university campus network to provide reliable access during the job fair.$$\nI shall develop Company Side (Web App) which includes following modules:\r\n\r\nSign Up \/ Login: Secure company registration.\r\nQR Code Attendance Marking (Mandatory): Verify physical presence.\r\nConfirm Arrival: Ensure recruiters are present on campus.\r\nEnter Skills Required: Declare required skill sets.\r\nView Student Profiles\/CVs & Projects \u2013 Access detailed student portfolios.\r\nSend\/Receive Interview Requests: Manage interview opportunities.\r\nAccept\/Reject Requests: Approve or decline student interview requests.\r\nManage Interview Queue: Auto-managed scheduling system that prevents overlapping interviews; if a time clash occurs, the system automatically adjusts interview slots based on company and student availability.\r\nNotifications: Send and receive notifications from the admin or students for updates and \r\nannouncements.\r\nSubmit Feedback: Provide structured event feedback.\r\nGet Participation Certificates: Automatically receive digital participation certificates post-event.\r\nLogout: Secure session exit.\r\nHosting and Deployment: The company module will be hosted and deployed on the university campus network to provide reliable access during the job fair.$$\nI shall develop Admin Side (Web App) which includes following modules:\r\n\r\nSign Up \/ Login: Authorized access for event admins(Career Development Center).\r\nMonitor Company Attendance: Track physically arrived companies.\r\nView Company & Student Details: Centralized event dashboard.\r\nNotifications: Send and receive important announcements or updates to\/from students and companies.\r\nAuto-Allocate Rooms: Assign interview rooms automatically.\r\nView Analytics: Statistics on interviews, hires, and participation.\r\nAnalyze Feedback: Collect and review company feedback.\r\nView Interview Status (Hired\/Rejected): Transparency in outcomes.\r\nMaintain Semester-Wise Data: Stores records of each semester\u2019s job fair, including student and company participation, hiring outcomes, and key statistics. Can be extended to include alumni data for future use in meetups, networking, and tracking career progress. Data can be shared with authorized personnel.\r\nLogout: Secure exit from system.\r\nHosting and Deployment: The admin module will be hosted and deployed on the university campus network to provide reliable access during the job fair.$$\n$$\n$$\n$$\n$$\n","comments":" $$ The revised proposal shows clear progress in addressing earlier feedback, adding depth with deployment, alerts, and semester-wise management. The system architecture is now well-organized and technically feasible. Still, integration with external professional platforms like LinkedIn and automated alumni tracking needs further definition. Overall, it\u2019s a strong and well-structured project ready for acceptance with minor clarifications.","isDraft":0,"status":2,"created_at":"2025-10-08 15:34:30","updated_at":"2025-10-17 11:39:37","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1040,"project_id":1376,"title":"Bridal Ease \u2013 A Smart Bridal Dress Renting and Selling Platform for Personalized Bridal Experience","prob":"In Pakistan, bridal dresses are among the most expensive wedding expenses, yet they are usually worn only once, making them a huge financial strain for families. Brides often face confusion and pressure while deciding whether to buy or rent because affordable and reliable options are scarce. Most must visit multiple physical stores, spending weeks searching and negotiating and still facing uncertain quality, authenticity, and pricing. The process is stressful, time-consuming and lacks transparency, with no verified reviews or standardized information to guide decisions. Existing wedding platforms mostly highlight venues and event services while completely overlooking the biggest challenge many brides face finding a beautiful bridal dress that is both accessible and affordable. This gap leaves countless brides without dependable rental choices, forcing them to compromise on design, fit, or price and turning what should be an enjoyable milestone into an exhausting, costly, and uncertain search for the perfect wedding dress.","description":"Bridal Ease \u2013 A Smart Bridal Dress Renting and Selling Platform for Personalized Bridal Experience is an intelligent and user-friendly system designed to simplify the bridal dress selection, rental, and purchase process. The platform enables brides to create secure profiles, explore a complete catalog of dresses with detailed descriptions, and make informed rental or purchase decisions, while also providing personalized recommendations through intuitive filtering and user-driven preferences.The system provides a filter-based and interactive dress catalog, allowing brides to search and sort dresses by budget, color, size, fabric, style, and occasion. A virtual try-on feature allows brides to visualize dresses on pre-built digital avatars, customized according to their body measurements and skin tone, before confirming a booking, enhancing confidence and reducing indecision.Bridal Ease efficiently manages rentals, deposits, and payments through a structured backend, ensuring transparency and smooth operations. The platform tracks rental history, dress availability, and popular trends using simple analytics, helping users make confident choices. A budget assistance module guides brides in choosing the most cost-effective option between renting or purchasing dresses, while vendors can easily manage inventory and receive real-time notifications.The frontend is developed using Flutter, ensuring a responsive and seamless experience across mobile devices, while the backend uses Supabase for secure datamanagement, authentication, and storage. The platform also includes a vendor verification process, an authentic review system, and simple administrative tools to ensure trust, accountabilit and quality control.\r\n\r\nBy focusing on bridal dress rentals and purchases, Bridal Ease reduces the financial burden of weddings, minimizes time spent searching through multiple stores, and promotes sustainability through shared use of luxury garments. The project ultimately transforms the traditional, stressful bridal shopping process into a smart, efficient, and enjoyable digital experience, empowering brides with convenience, affordability, and confidence while fostering a reliable community of vendors and satisfied users$$\nUser Management Module:This module handles registration, authentication, and role-based access for brides, vendors, and admins. User details, hashed passwords roles, and profile info are stored in Supabase tables. Rental history, wishlists, and preferences are linked by user_id.Login verification generates authentication tokens for secure sessions, and session management tracks active users, enabling personalized interactions. It also supports password recovery, multi-factor authentication, and profile updates.\r\nDress Catalog Module: Vendors upload dresses with metadata such as color, size, fabric, style, event, season, and price. Images are stored in Supabase, metadata in relational tables. Users can search, filter, and sort dresses. Supabase efficiently fetches results, allowing dynamic updates and real-time catalog browsing. The catalog supports category grouping, tag-based search, and metadata-driven sorting. Users can bookmark favorites and compare dresses.\r\nRecommender Module: Recommendations are generated using user preferences, browsing history, and wishlist activity stored in Supabase. Interaction data is updated when a user views, saves, or rents a dress. The system queries these tables and applies filter-based logic to suggest relevant dresses, providing a semi-personalized guided shopping experience. It tracks frequently paired items, offers alternative suggestions, and highlights similar styles.\r\nRental Management Module: Manages the rental lifecycle. Availability is verified in Supabase to prevent double bookings. Rental records store user ID, dress ID, start\/end dates, deposit, and status. Payments update rental status, and returned dresses restore availability. Notifications inform users about rental dates, returns, and changes. Cancellation rules, extensions, and late fees are handled. Analytics track frequent rental patterns.\r\n\r\nBudget Assistance Module: Guides brides in cost-effective choices between renting and buying. Rental and purchase prices from Supabase are compared with user budget preferences. Calculations highlight optimal financial decisions in the Flutter UI and suggest alternatives if a dress exceeds the budget. Discounts, seasonal offers, and package deals are considered.\r\n\r\nDress Popularity Tracker Module: Tracks rentals, views, and wishlist additions. Counters in Supabase are incremented per interaction. Popular dresses are displayed to users, while vendors can plan inventory based on demand. Historical analytics show seasonal trends, peak rental periods, and trending styles.\r\n\r\nVirtual Try-On And Avatar Module: Brides visualize dresses on a 3D avatar customized to body measurements and skin tone. Avatar settings and dress links are stored in Supabase. Flutter retrieves data to render realistic previews, allowing rotation and zoom to evaluate fit, style, and fabric appearance. Users can save snapshots, share previews, and receive feedback.\r\n\r\nPayment and Delivery Module: Manages secure transactions and delivery tracking. Payment records, including rental ID, amount, status, and gateway references, are stored in Supabase, along with delivery details. Real-time updates provide transparency. Refunds, disputes, and automated receipts are supported. The module integrates with courier services for scheduling returns and notifications.\r\n\r\nAdmin Panel Module: Provides administrators full control. Admins can verify vendor uploads, manage disputes, approve listings, and monitor rentals, payments, and popular dresses. Row-level security ensures only admins can access sensitive data. Dashboards display analytics, trends, and revenue reports, while direct database updates maintain platform integrity. Admins can generate the reports, send notifications, and manage platform performance efficiently.$$\nStudent-1 will manage the backend and AI-related functionality of Bridal Ease, focusing on the User Management, Rental Management, Recommender System, Virtual Try-On & Avatar, and Payment and Delivery modules. They will implement secure user authentication with role-based access, efficiently handle the rental operations including bookings, deposits, and returns, and ensure accurate, consistent, and secure data storage in a Supabase. The Recommender System will provide the semi-personalized suggestions based on user preferences, browsing history, and budget. For Virtual Try-On, a 3D avatar will dynamically adapt to user body measurements, shape, and skin tone, providing realistic, interactive, and engaging previews. Student-1 will also integrate the secure payment through gateways, manage delivery tracking, and ensure the seamless communication between frontend and the backend. Their focus is on backend stability and robust data handling, efficient AI logic, database integrity, and delivering a smooth, scalable, and reliable digital experience across the multiple devices successfully completed.$$\nStudent-2 will handle frontend development and catalog management, focusing on the Dress Catalog, Admin Panel, and Dress Popularity Tracker modules. They will design a responsive, intuitive, and visually engaging interface for brides to browse the dresses with the high-quality images, metadata, detailed descriptions, prices, availability, and special offers. The Dress Catalog will allow the filtering and sorting by size, color, fabric, style, season, and the budget for smooth navigation and enhanced discoverability. Through the Admin Panel, Student-2 will facilitate efficient management of users, dress listings, and bookings with real-time updates, notifications, and analytics. The Popularity Tracker will analyze rental trends, wishlist activity, and user engagement to display trending dresses, guiding user choices effectively. Student-2 will ensure the design consistency, performance optimization, seamless integration with backend APIs, and responsive layouts, delivering an interactive, reliable, and visually appealing user experience that enhances accessibility, satisfaction, engagement, overall platform usability, and user retention successfully.$$\n$$\n$$\n$$\n$$\n","comments":" $$ it is accepted after satisfactory revision with only slight refinements needed in AI training and model validation stages.","isDraft":0,"status":2,"created_at":"2025-10-22 13:09:48","updated_at":"2025-10-27 14:33:39","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1018,"project_id":1395,"title":"AutoFix: Intelligent Multi-Branch Vehicle Service Hub","prob":"Multi-branch vehicle service centers operating across cities such as Taxila, Wah Cantt, Hasanabdal, and Rawalpindi struggle with inefficient booking systems, leading to customer frustration and lost revenue. Customers face unclear service pricing, lack real-time updates, and receive no personalized maintenance guidance. They also struggle to identify the exact maintenance their vehicle needs, often relying on guesswork -highlighting the need for an AI-powered system that predicts required services proactively .Managers cannot effectively track mechanic workloads or monitor inventory levels across branches, causing service delays. Poor communication between branches results in inconsistent service quality and missed business opportunities. Traditional manual processes create data silos, preventing analytical insights for business optimization. These inefficiencies reduce customer satisfaction, increase operational costs, and limit the growth potential of service centers.","description":"AutoFix is a web-based platform for multi-branch vehicle service centers to manage bookings, services, and spare parts efficiently. Customers can book appointments in real-time, view service durations and prices, interact with an AI assistant for guidance, and provide feedback.By entering vehicle details, users receive AI-driven predictions on whether maintenance is needed and which specific service is required, enabling proactive and data-driven care. Managers can oversee branch operations, track mechanic workload, manage inventories, and generate analytical reports. Integrated notifications and activity monitoring enhance transparency, user engagement, and operational efficiency across the platform. By centralizing all operations, AutoFix improves customer experience, ensures proper inventory management, and provides predictive insights for service optimization.$$\nModule 1: User Authentication & Profile Management\r\nSecure JWT-based login\/signup with role-based access for Customer, Manager, and Admin. Includes profile management for multiple vehicles.\r\nModule 2: Service Booking & Appointment System\r\nReal-time booking with calendar integration, slot availability checking, branch working hours validation, and dashboard for service overview.\r\nModule 3: Multi-Branch Service & Inventory Management\r\nManages branch profiles, staff allocation, services offered, spare parts tracking across branches, and automated low-stock notifications.\r\nModule 4: AI Customer Assistant & Recommendation Engine\r\nThis module uses a Random Forest Classifier trained on a 50,000-record Vehicle Maintenance dataset https:\/\/github.com\/ahmedbutt53\/Vehicle-Maintenance-Dataset\r\nto predict whether a car needs maintenance and identify the exact service required. Users enter vehicle details like mileage, fuel efficiency, and component condition, and the system provides accurate, data-driven maintenance recommendations. Chatbot enhances user experience by offering personalized responses, cost estimates, and service duration forecasts, making the process proactive and user-friendly.\r\nModule 5: Vehicle Service Tracking & Allocation\r\nService history logs, quality tracking, mechanic workload management, and task allocation system.\r\nModule 6: Centralized Analytics & Multi-Dashboard System\r\nReal-time analytics for bookings, inventory, and revenue with branch-wise KPI dashboards, data visualization, and exportable reports.\r\nModule 7:Notification & Activity Monitoring System\r\nStreamline operations with instant notifications for bookings, services, and inventory, plus real-time activity tracking for transparency and efficiency\r\nModule 8: Customer Feedback & Blogs\r\nRating and review collection system, testimonials management, and educational blogs for car care tips and service awareness.$$\nModule 1 : Service Booking & Appointment System\r\nModule 3 : AI Customer Assistant & Recommendation Engine\r\nModule 5: Centralized Analytics & Multi-Dashboard System\r\nModule 7: Customer Feedback & Blogs$$\nDefault Module : User Authentication & Profile Management\r\nModule 2: Multi-Branch Service & Inventory Management\r\nModule 4: Vehicle Service Tracking & Allocation\r\nModule 6: Notification & Activity Monitoring System$$\n$$\nSpare Parts E-commerce App$$\nAI-powered Customer Assistance & Predictive Recommendations$$\nCentralized Analytics Dashboards with Data Visualization$$\nIntegrated Multi-Branch Inventory Management with Alerts","comments":" $$ The students have effectively addressed all committee concerns. The revised proposal clearly defines the AI prediction mechanism, including dataset source, algorithm (Random Forest Classifier), and prediction parameters. The workflow and data availability issues are now resolved. The AI and analytics components are better integrated into the system\u2019s core functionality, showing both technical feasibility and domain understanding.","isDraft":0,"status":2,"created_at":"2025-10-08 22:30:02","updated_at":"2025-10-17 11:38:41","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1034,"project_id":1362,"title":"AI-Driven Smart Enterprise Resource Planning (ERP) System for TradeWell","prob":"TradeWell, an importer and exporter of telecom equipment, faces challenges due to a lack of integration between its inventory, purchasing, and customer management systems. This causes data duplication, manual errors, and poor coordination across departments. The proposed Intelligent ERP System will centralize all business operations on one platform. It will also include an AI-based analytics module to forecast product demand and suggest optimal order quantities. This will improve efficiency, accuracy, and data-driven decision-making for the company.","description":"The Intelligent ERP System for TradeWell is a comprehensive web-based solution that integrates and automates the company\u2019s core business processes. It combines inventory management, purchase order handling, customer relationship management, and reporting into a single centralized system, ensuring real-time data flow and improved coordination across departments. The system also introduces an AI-based Predictive Analytics Module that analyzes historical sales and order data to forecast product demand and recommend optimal order quantities for various telecom products. This automation minimizes manual errors, reduces delays in procurement, and enhances overall operational efficiency. By providing data-driven insights through interactive dashboards, the system enables TradeWell to make informed decisions, maintain ideal stock levels, and strengthen its business management practices.$$\nThe Intelligent ERP System for TradeWell consists of five main modules.\r\n\u2022 User Management Module: Provides secure login and role-based access control for Admin, Manager, and Employee.\r\n\u2022 Inventory Management Module: Manages telecom product stock, updates quantities, and generates low-stock alerts.\r\n\u2022 Purchase Order and CRM Module: Handles supplier and customer details, creates purchase orders, and tracks payments and deliveries.\r\n\u2022 Reports and Analytics Module: Generates visual reports and dashboards for sales, inventory, and customer insights.\r\n\u2022 AI-Based Predictive Analytics Module: Forecasts product demand and recommends optimal order quantities using historical data.$$\nIn ERP, I shall develop the Inventory Management Module, which includes: (i) adding and updating telecom product stock records, (ii) managing stock-in and stock-out transactions, (iii) generating low-stock alerts, (iv) monitoring inventory status in real time, and (v) producing stock reports for management.$$\nIn ERP, I shall develop the Purchase Order and CRM Module, which includes: (i) managing supplier and customer details, (ii) creating and approving purchase orders, (iii) tracking order status and payments, (iv) handling customer feedback and complaints, and (v) implementing AI-based predictive analytics to forecast demand and suggest optimal order quantities.$$\n$$\nGeneral-purpose ERP system for small businesses.$$\n1. Integrated ERP with Inventory, Purchase, CRM, and AI modules.$$\n2. AI-based demand forecasting and order recommendation.$$\n3. Real-time analytics dashboard for business insights.","comments":" $$ Ensure all changes incorporated.","isDraft":0,"status":2,"created_at":"2025-10-19 15:58:43","updated_at":"2025-10-27 14:31:02","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1006,"project_id":1419,"title":"CUI Wah Job Fair Student Portal","prob":"Previously, the entire job fair process was manual, which create several inefficiencies for students. If student wanted to grab opportunity, they had to wait in long queues just to learn what company was offering. Students also had to carry multiple hard copies of their CVs and physically hand them out to each company an exhausting task, especially when there were dozens of recruiters.\r\nThere was no centralized platform for managing applications or interviews, so students couldn\u2019t easily track their progress or know which companies had shown interest. With this proposed system, all of these limitations are addressed digitally. Students will now\r\nhave centralized online portal where they can:\r\nCreate and manage their digital CV\/profile,\r\nRequest companies for interviews,\r\nReceive instant notifications for interview calls, and\r\nAccept or reject invitations directly through the app.\r\nThis digital transformation will make the entire job fair process\r\nfaster, transparent, and accessible for every student.","description":"This project focuses on developing the student-side portal of the CUI Wah Job Fair System, available as both a Flutter Web Application and Mobile Application. The system will allow students to:\r\n* Create their professional CV\/Profiles dynamically within the app.\r\n* Manage their interview slots with collision-free scheduling.\r\n* Receive real-time push notifications for interview calls or updates from companies.\r\n* Track interview status and view results.\r\n* View company details, job roles, and slot availability.\r\nThe app enhances student experience by providing a modern, interactive, and mobile-first interface while keeping their data secure and organized.$$\n1. Authentication Module \u2013 Secure authentication system allowing students to register and access their accounts and password recovery via email or OTP.\r\n2. Profile Creation \u2013 Students enter their personal, academic, and skill details to form a complete digital profile. Preview or export profile as PDF. Edit and update details anytime.\r\n3. CV Builder \u2013 A built-in tool to generate structured CVs automatically and dynamically within the app (education, experience, projects, skills) from entered data; stored data can later be used for analysis or matching with company requirements.\r\n4. Upload Final Year Project Details \u2013 Students can upload project titles, descriptions, and demo links for company review.\r\n5. View Companies and Required Skills \u2013 Allows students to browse all participating companies with details of desired skills and available positions.\r\n6. Company Insights Module \u2013 This module allows students to browse all registered companies participating in the job fair. They can view details such as available roles, required skills, and interview slot availability, helping them prepare and apply strategically based on their interests and qualifications.\r\n7. Send and Receive Interview Requests \u2013 Students can apply for interviews and also respond to incoming requests from companies.\r\n8. Accept or Reject Requests \u2013 Flexibility to accept or decline interviews based on preference or schedule.\r\n9. Auto-Managed Interview Schedule \u2013 This module allows students to view and book available interview slots with companies. It includes auto-collision detection to prevent overlapping interviews and automatically resolves timing conflicts. Students can also track their interview history, including accepted, pending, and rejected interviews, for better schedule management.\r\n10. View Interview Status \u2013 Students can track results such as hired, shortlisted, or rejected after interviews.\r\n11. Notifications \u2013This module enables students to receive real- time push notifications for important updates, including interview calls from companies, schedule changes or cancellations, and general announcements from the admin or participating companies, ensuring they stay informed throughout the event.\r\n12. Logout \u2013 Ends user session securely to protect account data.\r\n13. Hosting and Deployment \u2013 The student module will be hosted and deployed on the university campus network to provide reliable access during the job fair$$\nMuhammad Hassan Askari (CIIT\/FA22-BCS-155\/WAH) will independently develop the Student Module for the CUI Job Fair Student Portal, covering both mobile and web platforms using Flutter. His responsibilities include designing an intuitive and responsive interface that allows students to register, build digital CVs, view company details, schedule interviews, and receive real-time notifications. He will also develop the backend APIs using C# (ASP.NET) to ensure secure data handling and communication between frontend and database. The database will be implemented in PostgreSQL for efficient data management. Additional tasks include integrating OTP\/email authentication, implementing auto-collision detection for interview scheduling, and deploying the application on the university\u2019s campus network for stable access. His focus will be on providing a smooth, interactive, and secure experience for students, ensuring all features function seamlessly throughout the job fair event.$$\n$$\n$$\n$$\n$$\n$$\n","comments":" $$ The revised proposal shows clear progress in addressing earlier feedback, adding depth with deployment, alerts, and semester-wise management. The system architecture is now well-organized and technically feasible. Still, integration with external professional platforms like LinkedIn and automated alumni tracking needs further definition. Overall, it\u2019s a strong and well-structured project ready for acceptance with minor clarifications.","isDraft":0,"status":2,"created_at":"2025-10-08 15:28:48","updated_at":"2025-10-20 11:25:58","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1035,"project_id":1353,"title":"Smart Food Waste Reduction and Donation App","prob":"This project tackles the challenges of food waste and hunger. Enormous quantities of perfectly edible food are discarded daily by restaurants, hotels, and households, while underprivileged individuals struggle to secure even basic meals. Existing systems are fragmented, heavily manual, and lack smart regional coordination. \r\nOur proposed solution introduces a region-based mobile application that directly connects donors with receivers in real time. By leveraging geo-location and instant notifications, the system ensures surplus food is efficiently redistributed, reducing waste, combating hunger, and fostering a culture of sustainability and community resilience.","description":"1) The Smart Food Waste Reduction & Donation App creates a complete ecosystem for minimizing waste and fighting hunger. Donors register, add food details, and manage posts through the Donor Module, while receivers use the Receiver Module to view and request food. The Map Module displays available food with pins, and the Pick-up Point Module enables collection from shared community spots. The Food Expiry & Safety Module ensures only safe food is distributed, supported by the Notification Module for instant alerts. User accounts are handled by the User Management Module, with system oversight through the Admin Panel. The Feedback Module builds trust via ratings, while the Partnership Mode Module allows brands to run bulk donation campaigns. The Awareness & Advocacy Module educates users on reducing waste, and the Collaborative Campaign Module enables schools and organizations to run joint drives. Together, these modules create a smart, transparent, and socially impactful food-sharing platform.Voice-based UI Module An icon-based, text-free interface with Urdu\/English voice guidance for illiterate recipients.$$\nModules of the proposed System are:\r\n1) Donor Module 2) Receiver Module 3) Location & Map Module 4) Pick- up Point Module 5) Notification & Alert Module 6) User Management Module 7) Food Expiry & Safety Module 8) Feedback & Rating Module 9) Admin Control Panel 10) Partnership Mode module 11) Awareness & Advocacy Module 12) Collaborative Campaign Module 13) Voice-based UI Module$$\nDonor Module: Handles donor registration, adding\/updating food details, and managing donations.\r\nLocation & Map Module: Provides location auto-detection, displays food availability on maps, and allows region-based filtering. \r\nFood Expiry & Safety Module: Tracks expiry dates, ensures safe consumption, and generates alerts for unsafe food.\r\nNotification & Alert Module: Sends real-time alerts for new posts, reminders for expiry, and collection updates.\r\nPartnership Mode Module: Enables restaurants, brands, and organizations to run bulk food donation campaigns under their own name.\r\nAwareness & Advocacy Module: Educates users on the negative impacts of food waste and promotes sustainable consumption practices.\r\nVoice-based UI Module: Provides an easy-to-use, icon-based interface with multilingual voice instructions for illiterate or needy users.$$\nReceiver Module: Allows receivers to register, request food, and view available donations.\r\nPick-Up Point Module: Manages food collection spots, scheduling, and logistics.\r\nUser Management Module: Handles roles, profiles, and access control for all users.\r\nAdmin Control Panel: Provides overall system monitoring, donor\/receiver management, and reports.\r\nCollaborative Campaign Module: Enables schools and organizations to run joint food donation or awareness campaigns directly through the app.\r\nFeedback & Rating Module: Collects ratings and feedback from donors\/receivers for trust and improvement.$$\n$$\nHotel Food Donation Application$$\n1. Real-time Geo-location & Map Integration \u2013 donors can pin exact locations while receivers view nearby food through an interactive map.$$\n2.Smart Region-Based Notifications \u2013 receivers instantly get alerts for food posted in their area, ensuring timely collection$$\n3. Food Expiry & Safety Tracking \u2013 The system monitors expiry times of posted food, removes unsafe items automatically, and alerts receivers for soon-to-expire meals, ensuring safe","comments":" $$ The project scope has significantly expanded with new social, technical, and HCI-driven modules. The voice-based and campaign features make it both inclusive and scalable. Earlier concerns about usability and innovation have been comprehensively addressed.","isDraft":0,"status":2,"created_at":"2025-10-20 14:44:08","updated_at":"2025-10-27 14:32:30","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1012,"project_id":1379,"title":"Social Platform Exposure Tracker (SafeSocial)","prob":"Users on Facebook, YouTube, and Reddit often unintentionally expose personal data such as \r\nphone numbers and emails in posts, comments, and video descriptions. Additionally, \r\nadvertising platforms and third-party apps gain access to user activity, leading to targeted \r\nmanipulation, spam, scams, and identity theft. Our system identifies such exposures across \r\nplatforms and provides step-by-step security recommendations with AI chatbot support, \r\nhelping users strengthen their online privacy.","description":"The Social Platform Exposure Tracker is a web-based tool that scans Facebook, YouTube, \r\nand Reddit to detect where user\u2019s sensitive personal data is exposed in posts, comments, \r\nand video descriptions. It also analyzes ad insights from DIY data files (Facebook & \r\nYouTube) to show which companies and interests are being used for targeted advertising. \r\nUsers log in securely via OAuth or upload DIY data files. The system uses regex + NLP for \r\ndetection and provides step-by-step security recommendations. A conversational AI chatbot \r\nguides users in fixing issues and managing privacy. The system also cross-checks repeated \r\nexposures across platforms to highlight higher risks.$$\n\uf0b7 User Authentication Module \u2013 Secure login via OAuth for Facebook, YouTube, and \r\nReddit. \r\n\uf0b7 Data Fetch & Processing Module \u2013 Collects user posts, comments, and video metadata \r\nthrough APIs. \r\n\uf0b7 Exposure Detection Module \u2013 Identifies sensitive data using regex and NLP. \r\n\uf0b7 Ad Insights & Tracking Module \u2013 Processes DIY data (Facebook & YouTube) to show \r\nadvertisers, ad categories, and targeting. \r\n\uf0b7 Cross-Platform Risk Analyzer \u2013 Detects repeated exposures across multiple platforms. \r\n\uf0b7 Security Recommendations Module \u2013 Suggests platform-specific fixes. \r\n\uf0b7 Chatbot Module \u2013 Conversational assistant for privacy guidance. \r\n\uf0b7 Dashboard & Visualization Module \u2013 Displays results, risks, and recommendations \r\nclearly.$$\nIn this project I shall develop modules that are: \r\n- User Authentication & Input Handler (multi-platform) \r\n- Data Fetch & Processing Module \r\n- Ad Insights & Tracking Module \r\n- Security Recommendation Engine$$\nIn this project I shall develop modules that are: \r\n- Exposure Detection Engine (regex + NLP) \r\n- Cross-Platform Risk Analyzer \r\n- AI Chatbot Integration \r\n- Frontend Dashboard Development$$\n$$\n$$\n$$\n$$\n","comments":" $$ Most of the committee\u2019s concerns have been addressed. The scope has been expanded with the inclusion of YouTube and an additional Dashboard & Visualization Module. The dependency on DIY data has been partially reduced by integrating OAuth-based API data fetching, though data crawling or feasibility analysis across all platforms (especially Reddit) still needs more clarity.","isDraft":0,"status":2,"created_at":"2025-10-08 20:36:42","updated_at":"2025-10-17 09:52:52","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1032,"project_id":1364,"title":"The Digital Scholar","prob":"Our project focuses on developing a system that extracts information from fatwa collections and provides direct answers to user queries, making the search process faster and more convenient.","description":"Our project, The Digital Scholar, is an AI-powered mobile application designed to make authentic Islamic rulings digitally accessible through the comprehensive Fatawa Razawiya. The main objective of this project is to digitize, organize, and modernize access to this vast knowledge base so that users can ask questions in natural language and receive contextually relevant, authenticated answers from the digital archive.\r\n\r\nUnlike existing fatwa apps that rely on limited datasets or keyword-based search, our system will use advanced Natural Language Processing (NLP) to interpret user intent, understand query context, and retrieve precise answers. Since no digital dataset currently exists for Fatawa Razawiya, one of our key technical contributions will be creating a structured and searchable version of this text. This process includes text extraction, cleaning, annotation, and the addition of metadata such as topic classification, scholar references, and source indexing, ensuring that the information is both reliable and efficiently retrievable.\r\n\r\nAlthough the system initially focuses on Hanafi fiqh, Fatawa Razawiya itself covers a remarkably broad range of jurisprudential and ethical topics, thereby ensuring a wide and diverse project scope. Furthermore, the system architecture is designed to be scalable, enabling future integration of other fiqh datasets or modules without altering the core design.\r\n\r\nTo enhance accessibility, the project also introduces Automatic Answer Summarization for concise comprehension and Text-to-Audio Conversion for convenient listening. By combining classical Islamic scholarship with modern AI techniques, this project contributes both technically and socially by preserving, structuring, and intelligently presenting authentic religious knowledge in a digital form.$$\nMobile Application: Provides a cross-platform app interface enabling users to easily access, query, and view fatwa answers.\r\nVoice Interaction Module: Enables complete voice-to-voice interaction by converting system responses into natural speech using Text-to-Speech (TTS) technology, allowing users to communicate entirely through voice without typing or reading.\r\nUser Query Interface: A simple and user-friendly interface where users can input queries.\r\nAnswer Presentation Module: Displays the extracted answers along with the source link (if needed).\r\nAudio Q&A Module: Users can ask questions via voice, system transcribes and answers.\r\nQuery Processing & Interpretation: Understands and interprets user queries using Natural Language Processing (NLP) techniques.\r\nDatabase Management: Stores, organizes, and manages fatwa collections and related metadata.\r\nInformation Retrieval Engine: Fetches relevant content from the database that matches the interpreted query.\r\nAnswer Extraction & Generation: Identifies the most accurate answer from retrieved content, ensuring relevance and correctness.$$\nStudent 01, Maaz Abdullah will work on backend modules of the system.Following are the backend Modules\r\nBack-End Modules\r\n1.\tQuery Processing & Interpretation \u2013 Understands and interprets user queries using Natural Language Processing (NLP) techniques.\r\n2.\tDatabase Management Module \u2013 Stores, organizes, and manages fatwa collections and related metadata.\r\n3.\tInformation Retrieval Engine \u2013 Fetches relevant content from the database that matches the interpreted query.\r\n4.\tAnswer Extraction & Generation \u2013 Identifies the most accurate answer from retrieved content, ensuring relevance and correctness.$$\nStudent 02, Hanzala Shafique will work on Frontend modules of the system.Following are frontend Modules\r\nFront-End Modules\r\n1.\tMobile Application: Provides a cross-platform app interface enabling users to easily access, query, and view fatwa answers.\r\n2. User Query Interface: A simple and user-friendly interface where users can input queries.\r\n3. Answer Presentation Module: Displays the extracted answers along with the source link (if needed).\r\n4. Audio Q&A Module: Users can ask questions via voice, system transcribes and answers.\r\n5. Voice Interaction Module: Enables complete voice-to-voice interaction by converting system responses into natural speech using Text-to-Speech (TTS) technology, allowing users to communicate entirely through voice without typing or reading.$$\n$$\n$$\nDigitization and Structuring of Fatawa Razawiya (30 Volumes) \u2014 Development of a machine-readable and searchable dataset from authentic Hanafi fatwas.$$\nVoice Interaction Module: Enables complete voice-to-voice interaction by converting system responses into natural speech using Text-to-Speech (TTS) technology, allowing users to communicate entirely$$\nAutomatic summarization of answers for concise understanding.","comments":" $$ The project now shows strong technical depth with real contribution through dataset creation, voice-based modules, and summarization features. Scope has been sufficiently broadened by focusing on a full-scale digitization of Fatawa Razawiya. The NLP integration is now better justified and context-aware.","isDraft":0,"status":2,"created_at":"2025-10-17 18:34:32","updated_at":"2025-10-27 14:31:40","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":957,"project_id":1371,"title":"Web Application Security Scanner","prob":"Modern web applications are highly vulnerable to security threats such as injection attacks, broken authentication, misconfigurations, and other risks highlighted in the OWASP Top 10. Although there are commercial scanners available, many organizations especially small and medium enterprises lack affordable and accessible automated tools.\r\n\r\nExisting open-source tools (like Vulneradar, Nikto, Wapiti, and Arachni) either require technical expertise, have steep learning curves, or lack real-time feedback and risk-based recommendations. Developers and non-security experts struggle to quickly identify vulnerabilities, understand their severity, and take immediate action.\r\n\r\nThis project addresses this gap by creating a dual-mode Web Application Security Scanner that provides both:\r\n\r\nSoft Scan: Simplified and safe scanning option for layman users and developers.\r\nDeep Scan: Advanced scanning with detailed testing for experts.\r\n\r\nThe scanner also introduces risk-based recommendations, real-time reporting, and detection of rate-limiting & brute-force protections, making it more useful than existing lightweight scanners.","description":"A web-based application that scans websites for vulnerabilities, identifies underlying technologies, provides risk scores, and suggests fixes. It offers two scanning modes (Soft & Deep) and generates professional reports.\r\n\r\nHow:\r\n\r\ni) Users enter a target website URL.\r\nii) The system crawls and discovers endpoints.\r\niii) It identifies technologies (server, CMS, frameworks, libraries).\r\niv) Runs vulnerability tests (OWASP Top 10 + rate-limiting checks).\r\nv) Calculates risk scores.\r\nvi) Provides actionable remediation guidance.\r\nvii) Displays real-time results on a dashboard and stores scan history.\r\n\r\nObjectives:\r\n\r\ni)Provide an easy-to-use security scanner for developers and organizations.\r\nii) Support both soft scan (safe and quick) and deep scan (detailed and aggressive).\r\niii) Automate detection of OWASP Top 10 vulnerabilities.\r\niv) Enhance decision-making with risk-based recommendations.\r\nv)Improve awareness of outdated or insecure technologies.\r\nvi)Deliver comprehensive reports.$$\n(i) URL Input Interface\r\nUser enters the website URL for scanning.\r\n(ii) Endpoint Crawler\r\nAutomatically discovers and maps all accessible pages and links within the target website\r\n(iii) Technology Detection\r\nIdentifies the underlying web technologies, frameworks, and platforms used by the target website.\r\n(iv) Vulnerability Scanner\r\nTests for OWASP Top 10 (e.g., SQL injection using payload fuzzing, XSS with regex-based detection, CSRF token analysis, etc.).\r\n(v) Risk Assessment\r\nCalculates a risk score based on detected vulnerabilities and outdated technologies to prioritize issues.\r\n(vi) Fix Suggestions\r\nProvides tailored remediation recommendations based on the vulnerabilities and their risk levels.\r\n(vii) Soft Scan vs. Deep Scan \r\nSoft Scan: Limited to non-destructive checks (misconfiguration, outdated tech, headers, SSL\/TLS issues).\r\nDeep Scan: Includes injection testing, brute force checks, parameter tampering, and rate-limit bypass attempts.\r\n(viii) Report Generator\r\nGenerates detailed scan reports in JSON, PDF, and HTML formats for review and documentation.\r\n(ix) User Authentication\r\nImplements secure user registration and login with JWT-based authentication.\r\n(x) Scan History\r\nStores and manages past scan results for user reference and comparison.\r\n(xi) Live Scan Output\r\nDisplays real-time scanning progress and results using WebSockets for immediate feedback.$$\nIn Web Application Security Scanner, I shall develop module which include : (i) URL Input Interface (ii) Risk Assessment (iii) Report Generator (iv) Scan History$$\nIn Web Application Security Scanner, I shall develop module which include : (i) Endpoint Crawler (ii) Vulnerability Scanner (iii) Technology Detection (iv) Live Scan Output$$\n$$\nWebsite Pen Tester$$\n1. Develop Technology Detection module to identify web technologies used by the target site.$$\n2. Implement Risk-Based Recommendation module to suggest fixes based on vulnerability severity.$$\n3. Develop module to detect rate limiting and brute force protections.","comments":" $$ Ensure all changes incorporated in the revised version.","isDraft":0,"status":2,"created_at":"2025-09-29 11:59:58","updated_at":"2025-10-31 10:29:50","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1049,"project_id":1382,"title":"HealthInsura360","prob":"Health insurance is crucial for financial protection, yet many individuals struggle to understand complex policy details and choose the right plan. Existing online platforms often lack personalization and real-time support, leading to confusion and poor decision-making. Users face difficulties interpreting terms, comparing numerous plans, and managing multiple policies effectively. Limited awareness and lack of trusted guidance further discourage people from purchasing insurance. HealthInsura360 addresses these real-world challenges through Artificial Intelligence. The system features an AI-based Policy Recommendation Engine that analyzes user data to suggest suitable health plans and an AI Chatbot that provides instant, 24\/7 assistance to explain terms, answer questions, and guide users through the selection process. By combining intelligent recommendations with interactive support, HealthInsura360 simplifies policy understanding, enhances accessibility, and empowers users to make informed health insurance decisions confidently.","description":"HealthInsura360 is an AI-powered web-based health insurance management system designed to simplify and digitalize the process of purchasing, managing, and understanding health insurance. The system addresses key real-world challenges such as lack of policy awareness, difficulty in comparing plans, and limited personalization by integrating automation, artificial intelligence, and secure online operations.\r\nThe platform consists of multiple interconnected modules: User, Agent, Admin, Hospital, Payment, AI, Reporting & Analytics, and Authentication.\r\nUsers can register, verify their identity using OTP, and log in securely through JWT-based authentication. They can explore insurance plans, receive AI-driven policy recommendations, and interact with an AI chatbot for real-time assistance. Agent approve claims and verifies while the Admin oversees users, agents, hospitals, policies, and claims.\r\nA dedicated Hospital Module enables hospitals to submit insurance claims, upload treatment documents, and track claim status, ensuring a transparent and efficient process. The Payment Module integrates with test-mode gateways for secure premium payments and claim settlements. The Reporting & Analytics Module provides insights into user engagement, policy performance, and claim trends, assisting in data-driven decision-making.\r\nThe AI Module enhances the system\u2019s intelligence through personalized policy suggestions and a conversational chatbot, improving accessibility and user satisfaction.$$\n1.User Module:\r\nRegister and log in.\r\nUpdate personal profile and family information.\r\nBrowse, compare, and view policy details.\r\nGet AI-based policy recommendations based on personal data.\r\nInteract with the AI chatbot for instant support and queries.\r\nPurchase or renew health insurance policies.\r\nMake online payments.\r\nReceive notifications about renewals, offers, and policy updates.\r\nView and download policy documents and payment receipts.\r\n\r\n2.Agent Module:\r\nRegister and log in as an authorized insurance agent.\r\nView assigned customers and manage their policy applications.\r\nApprove claims and Follow ups.\r\nApprove or reject claim requests submitted by hospitals.\r\nTrack commissions.\r\n\r\n3.Admin Module:\r\nAdmin authentication and secure access using JWT.\r\nManage user and agent ,hospital accounts.\r\nManage hospital records and verify partnerships.\r\nAdd, update, and delete health insurance policies.\r\nGenerate analytical and financial reports.\r\nManage payment records and transaction history.\r\nReview Agent Commisions.\r\nAudit Logs\r\nHandle Complains\r\n\r\n4.AI Module:\r\nAI chatbot\r\nAI policy recommendation\r\n\r\n5.Payment Module:\r\nTracks payment history and generates receipts.\r\nIntegrates with payment gateways.\r\nEnables online premium payments and renewals.\r\nProvides refund and transaction status updates.\r\nSecurely process premium payments made by users.\r\nRecord transaction details and generate receipts.\r\nMaintain transaction logs for admin analysis.\r\nManage simulated payouts to hospitals for approved claims.\r\nDisplay payment status.\r\n\r\n6.Reports & Analytics Module:\r\nGenerate visual reports on sales, user growth, and agent performance.\r\nAnalyze AI recommendation accuracy and chatbot usage statistics.\r\nTrack payment trends and policy popularity.\r\nExport reports.\r\n\r\n7.Hospital Module\r\nHospital registration and admin approval.\r\nSecure hospital login and dashboard access.\r\nSearch for user policies using policy number or ID.\r\nSubmit claim requests for insured patients.\r\nUpload medical bills, diagnosis, and treatment details.\r\nTrack claim status (Pending \u2192 Approved \u2192 Paid).\r\nReceive notifications for payment updates.\r\nGenerate claim summary reports for internal use.\r\n\r\n8.Authentication Module:\r\nUses JWT (JSON Web Token) for secure user and admin sessions.\r\nValidates credentials during login and protects against unauthorized access.\r\nSupports role-based authentication (User \/ Agent \/ Admin\/ Hospital ).\r\nManages password resets and account recovery.\r\nLogs login attempts for security tracking.$$\n1. Agent Module\r\n\r\nEnables agents to assist users in selecting suitable insurance policies and managing client applications.\r\nAgents can view user details, track policy sales, and coordinate between users and the admin.\r\n\r\n2. Admin Module\r\n\r\nProvides full control over the system, allowing management of users, agents, hospitals, and insurance policies.\r\nThe admin verifies claims, monitors payments, and generates analytical reports for decision-making.\r\n\r\n3. Reports & Analytics Module\r\n\r\nGenerates detailed reports on sales, claims, user engagement, and agent performance.\r\nHelps the admin make data-driven decisions through visual charts and trend analysis.\r\n\r\n4. AI Module\r\n\r\nIncorporates an AI Chatbot for real-time assistance and an AI Recommendation System for personalized policy suggestions.\r\nEnhances user experience by providing intelligent, data-based guidance.$$\n1. User Module\r\n\r\nAllows users to register, explore policies, receive AI-based recommendations, and interact with the chatbot.\r\nUsers can buy, renew, or track policies and make secure online payments.\r\n\r\n2. Hospital Module\r\n\r\nEnables hospitals to register, submit claims, upload medical documents, and track payment status.\r\nEnsures a smooth claim settlement process between hospitals and the insurance system.\r\n\r\n3. Payment Module\r\n\r\nHandles all premium payments and claim settlements through payment gateways.\r\nRecords transaction history and generates digital receipts for transparency.\r\n\r\n4. Authentication Module\r\n\r\nUses JWT for secure, role-based login sessions for users, agents, admins, and hospitals.\r\nEnsures data privacy through encrypted credentials, OTP verification, and access control.$$\n$$\n$$\n1.\tHospital Module: Enables hospitals to register, submit patient claims, upload medical reports, and track claim approvals and payments securely.$$\n2.\tAI Module: Integrates an AI chatbot for real-time assistance and an AI-based policy recommendation system for personalized insurance suggestions.$$\n3.\tAuthentication Module: Implements JWT-based, role-specific login for secure access, session management, and activity tracking across all user types.","comments":" $$ Ensure the proposed AI model integration","isDraft":0,"status":2,"created_at":"2025-11-11 10:08:36","updated_at":"2025-11-21 12:03:10","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":960,"project_id":1340,"title":"DeepHunt AI \u2013 \u201cSmarter search, Bigger Opportunities.\u201d","prob":"Career exploration is long and frustrating. Research shows job seekers spend over 5 months and more than 10 hours per week searching for opportunities, with many reporting searches lasting 6+ months. This leads to missed chances, wasted effort, and poor career planning.\r\nExisting solutions are fragmented, focusing only on one domain (like scholarships or jobs), often restricted to specific regions, and lacking intelligent personalization.\r\nOur project solves this by creating DeepHunt AI, a multi-agent AI platform that continuously scans global and local sources for jobs, internships, and scholarships. It uses deep personalization with LLMs to match opportunities with user profiles, detect skill gaps, and recommend learning paths. It also assists in generating customized resumes and cover letters.","description":"DeepHunt AI is a next-generation platform for opportunity discovery. Users create detailed profiles (skills, academics, goals). Specialized AI agents then continuously scan the web for jobs, scholarships, and internships. A personalization engine powered by LLMs matches opportunities with profiles, identifies skill gaps, and recommends courses. A chatbot assistant enables conversational search, while the Smart Application Assistant generates tailored resumes and cover letters.\r\nDeepHunt AI\u2019s Multi-Agent Scanner will use a mix of APIs, RSS feeds, and web scraping where permitted.\r\nAPIs & Feeds: Preferred sources (Adzuna, Jooble, USAJobs, government portals, university RSS).\r\nStatic Sites: Scraped using requests + BeautifulSoup.\r\nDynamic Sites: Handled with Playwright\/Selenium for JavaScript-rendered content.\r\nAJAX \/ Infinite Scroll: Extracted via JSON\/XHR endpoints or controlled scrolling.\r\nForm-based Pages: Automated POST requests or simulated form submission if allowed.\r\nAll agents will respect robots.txt, ToS, and rate limits. If scraping is not allowed, alternatives like APIs,download and scrape, or user-imports will be used. Data is cleaned, normalized, and stored for AI-based personalization and ranking.$$\n1.\r\nUser Authentication & Profile Management :Secure signup\/login, profile creation with academics, skills, goals, and certifications.\r\n2.\r\nMulti-Agent Opportunity Scanner : Specialized agents (job, scholarship, freelance, conference) continuously scrape and filter results.\r\n3.\r\nAI Personalization Engine : LLM-based matching, skill-gap detection, and learning path suggestions.\r\n4.\r\nInteractive Dashboard & Analytics : Unified dashboard for results, notifications, career insights, and trends visualization.\r\n5.\r\nAI Career Assistant (Chatbot) : Conversational interface for career queries, resume\/cover letter generation.\r\n6.\r\nSmart Application Assistant : Manages applications, deadlines, and auto-generates tailored application material$$\nZubair Shafi (FA22-BCS-003)\r\nResponsible for backend architecture planning, modular service design, database schema creation, optimization and indexing, API development, integration workflows, and complete UI\/UX design focusing on user-friendly interfaces, accessibility standards, prototyping, usability testing, and iteration.$$\nHaseeb Ahmed (FA22-BCS-097)\r\nResponsible for multi-agent scraping pipelines, intelligent data extraction, LLM-powered personalization engine, fine-tuned recommendation modeling, adaptive matching algorithms, and comprehensive frontend development including interactive design, responsive layouts, testing, debugging, and deployment.$$\n$$\nCareer Assist : Institute recommendation system for unimpaired\/impaired students.\r\nScholarship Spy :Scholarship recommendation system with content-based filtering.\r\nCareer Guide (Hybrid App) : Flutter-based app for aptitude$$\n1.\r\nMulti-Domain Agentic AI : Unlike single-domain projects, Career Catalyst AI covers jobs, scholarships, internships, conferences, and freelance opportunities in one platform$$\n2.\r\nLLM-Powered Deep Personalization : Goes beyond simple recommendation by detecting skill gaps, generating adaptive learning paths, and providing personalized insights.$$\n3.\r\nSmart Application Support : Unique feature that auto-generates resumes\/cover letters, tracks applications, and sends deadline alerts functionality missing in prior projects","comments":" $$ Students have list of webiste for crawling and also tool which will be used for scraping. All the use cases should be implemented as proposed in proposal.","isDraft":0,"status":2,"created_at":"2025-09-29 21:33:20","updated_at":"2025-10-15 15:19:54","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1028,"project_id":1352,"title":"Skinalytics: The Intelligent Path to Better Skin","prob":"A significant number of individuals struggle with common skin concerns such as acne, pigmentation, and dryness, yet they face substantial barriers in accessing professional dermatological care due to high costs, limited availability, and long wait times. This lack of accessible expertise forces people to self-treat through guesswork, generalized online advice, or ineffective over-the-counter products. Consequently, they often invest in unsuitable routines that fail to address their specific needs and can potentially aggravate their skin conditions. This creates a widespread need for a reliable, immediate, and affordable solution to provide personalized skincare guidance directly to consumers.","description":"Skinalytics is an advanced AI-powered mobile application designed to democratize personalized skincare. By simply uploading a selfie, users receive an instant, intelligent analysis of their skin, detecting concerns like acne, wrinkles, pigmentation, and dryness. The platform then generates a complete, personalized skincare regimen, including structured daily and weekly routines and a dermatologist-style PDF report.\r\n\r\nThis process is integrated with a Virtual Skincare Store, where users can explore and purchase recommended products. The overall experience is further supported by a robust Customer Relationship Management (CRM) System, which is not an online store, but a personalization and engagement backbone connecting the store and user dashboard. The CRM continuously tracks user profiles, analysis results, purchase history, and progress to refine recommendations, send reminders, and manage user interactions. It also helps administrators analyze user behavior, satisfaction trends, and engagement performance, ensuring a seamless, data-driven, and user-focused\u00a0experience.\r\nIn short, the CRM ensures that every user\u2019s skincare journey \u2014 from diagnosis to ongoing care \u2014 remains personalized, connected, and adaptive.\r\n\r\nBuilt with privacy and scalability in mind, all sensitive facial data is handled in strict compliance with GDPR and Swiss data protection regulations, requiring explicit user consent. For rapid deployment, the app will leverage external APIs, with a strategic roadmap to develop a proprietary AI model for greater innovation and independence. Future extensions include real-time skin analysis via the device camera, creating a dynamic and highly differentiated user experience.$$\n(i) User Authentication & Consent Module\r\nThis module ensures secure user registration and login, along with profile management. It also manages user consent for processing sensitive facial data, ensuring compliance with GDPR and privacy regulations.\r\n\r\n(ii) AI-Powered Skin Analysis Engine\r\nThis is the core intelligence of the system, responsible for analyzing selfies to detect skin conditions such as acne, wrinkles, pigmentation, and dryness. It leverages advanced AI models to deliver accurate and reliable results.\r\n\r\n(iii) Personalized Routine Management Module\r\nBased on the analysis results, this module generates tailored daily and weekly skincare routines. It adapts to individual needs, providing structured and practical guidance for users.\r\n\r\n(iv) Virtual Skincare Store Module\r\nThis module integrates product exploration and purchase functionality. It allows users to directly browse and buy recommended skincare products aligned with their personalized routines.\r\n\r\n(v) Report Generation & Export Module\r\nUsers can receive professional dermatologist-style reports in PDF format. This module compiles skin assessments, routines, and progress into structured, downloadable documents for easy reference.\r\n\r\n(vi) User Dashboard & History Tracking Module\r\nThe dashboard provides a centralized view of user activity. It enables tracking of past analyses, progress monitoring, and quick access to reports and routines.\r\n\r\n(vii) Real-Time Skin Detection Module\r\nThis module uses the mobile camera for live skin analysis. It enables dynamic, interactive detection of skin concerns, creating an engaging and immediate user experience.\r\n\r\n(viii) Customer Relationship Management (CRM) Module\r\nManages user profiles, preferences, and progress while connecting data between the Virtual Store and User Dashboard. It personalizes experiences, sends reminders, and tracks engagement, ensuring seamless coordination across the system.$$\nI shall develop (i) User Authentication & Consent Module, (ii) Report Generation & Export Module, (iii) User Dashboard & History Tracking Module, and (iv) Customer Relationship Management Module, ensuring security, reporting, tracking, and user engagement.$$\nI shall develop (i) AI-Powered Skin Analysis Engine, (ii) Personalized Routine Management Module, (iii) Real-Time Skin Detection Module, and (iv) Virtual Skincare Store Module, covering analysis, personalization, detection, and product integration.$$\n$$\n1.\tSkin Disease Detection-SDD\t\r\n2.\tSkin Lesion Cancer Detection$$\n1. Proprietary AI Skin Analysis: This project goes beyond basic detection by using advanced AI to analyze selfies for detailed cosmetic concern with the insights directly feeding a centralized system.$$\n2. Personalized Routine & Virtual Store Integration: The system generates personalized skincare routines and integrates a virtual store for buying recommended products.$$\n3. Dashboard with Reports & History Tracking: A dedicated user dashboard stores past analyses, enables progress tracking, and provides downloadable structured PDF reports that mimic a dermatologist\u2019s.","comments":" $$ Ensure all changes incorporated in the revised version.","isDraft":0,"status":2,"created_at":"2025-10-16 08:13:36","updated_at":"2025-10-31 10:28:58","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1011,"project_id":1380,"title":"Smart Meal Scanner, Nutrition Estimator & Weight\r\nLoss Planner","prob":"Maintaining a healthy lifestyle has become a major challenge in today\u2019s fast-paced world. Many people either lack the time or the knowledge to monitor their diet and daily activities. Current health and fitness apps often require users to manually log every meal, which is time-consuming, repetitive, and discouraging. In addition, most recommendations provided are generic and fail to consider personal factors such as age, weight, height, allergies, and fitness goals. This lack of personalization not only reduces effectiveness but can also be risky for people with food allergies, since existing apps rarely warn users about unsafe meals. As a result, people struggle with incomplete tracking, low motivation, unsustainable results, and sometimes unsafe dietary choices. To overcome these issues, there is a strong need for a smart, user-friendly system that can automatically recognize meals, calculate nutrition, check for allergies, provide personalized plans, and log both diet and daily activities.","description":"The proposed project, Smart Meal Scanner, Nutrition Estimator & Weight Loss Planner, is a mobile-based application designed to help users manage their diet, exercise, and health more easily and safely. The main goal of this project is to provide a smart and personalized solution that automates meal tracking, estimates nutrition, detects food allergies, and generates customized diet and workout plans.\r\n\r\nThe system allows users to log their meals in two ways: by capturing food images or by manual entry. Using deep learning (CNN with MobileNetV2 or YOLOv5) trained on the Food-101 dataset, the application automatically recognizes food items from the image. Once recognized, the system retrieves accurate nutritional information\u2014such as calories, proteins, fats, and carbohydrates\u2014from the USDA FoodData Central API.\r\n\r\nThe application also considers the user\u2019s age, weight, height, gender, and allergies to provide a personalized experience. It checks each scanned meal against the stored allergy information and warns users about unsafe foods while suggesting safe alternatives. The system further uses BMR (Basal Metabolic Rate) and TDEE (Total Daily Energy Expenditure) formulas to design daily diet and workout plans according to user goals such as weight loss, gain, or maintenance.\r\n\r\nUsers can track their daily activities and calories burned, while the app automatically generates progress reports and visual charts to show improvements over time. An admin panel allows management of food data, exercises, and allergy information.$$\nUser Authentication Module:\r\nHandles secure user signup and login using Firebase Authentication. It allows users to save personal details such as age, weight, height, and goals to create a personalized profile.\r\n\r\nUser Profile Management Module:\r\nAllows users to update and manage their personal information, fitness goals, and dietary preferences. This information helps the system provide customized diet and workout recommendations.\r\n\r\nFood Recognition Module (AI-Based):\r\nUses a Convolutional Neural Network (CNN) model such as MobileNetV2 or YOLOv5, trained on the Food-101 dataset, to automatically identify food items from images taken by the user.\r\n\r\nImage Preprocessing Module:\r\nImproves the accuracy of the AI model by resizing, enhancing, and normalizing food images before recognition. This step helps the model work more efficiently.\r\n\r\nNutrient Estimation Module:\r\nAfter recognizing the food, this module calculates nutrition details like calories, proteins, fats, and carbohydrates using data from the USDA FoodData Central API.\r\n\r\nMeal Entry and History Module:\r\nAllows users to log their meals automatically (through recognition) or manually if needed. It also keeps a daily record of all meals for later tracking and comparison.\r\n\r\nExercise and Calorie Burn Module:\r\nRecommends workouts based on user fitness goals and calculates calories burned according to the intensity and duration of exercises.\r\n\r\nFood Allergy & Restriction Module:\r\nEnsures user safety by checking all meals against the user\u2019s stored allergy information. If an unsafe food is detected, it provides an alert and suggests a safe alternative.\r\n\r\nCustomized Diet Planning Module:\r\nGenerates personalized diet plans using a rule-based approach that calculates daily calorie needs through BMR (Basal Metabolic Rate) and TDEE (Total Daily Energy Expenditure) formulas. Plans are based on user age, weight, goals, and allergy data.\r\n\r\nDaily Activity Tracking Module:\r\nRecords physical activities such as walking, running, or workouts and estimates calories burned throughout the day. This helps balance calorie intake with calorie expenditure.\r\n\r\nReports & Visualization Module:\r\nShows progress through charts and graphs, comparing calories consumed versus calories burned. It also displays weekly and monthly reports to motivate users.\r\n\r\nAdmin Management Module:\r\nProvides an admin dashboard where food data, exercises, and allergy information can be updated. This ensures that the system remains scalable and easy to manage.$$\nI shall develop the following modules:\r\n\r\nUser Authentication Module\r\n\r\nUser Profile Management Module\r\n\r\nFood Recognition Module (CNN \u2013 MobileNetV2\/YOLOv5, Food-101 dataset)\r\n\r\nImage Preprocessing Module\r\n\r\nNutrient Estimation Module\r\n\r\nMeal Entry and History Module$$\nI shall develop the following modules:\r\n\r\nExercise and Calorie Burn Module\r\n\r\nFood Allergy & Restriction Module\r\n\r\nCustomized Diet Planning Module (Rule-based using BMR and TDEE)\r\n\r\nDaily Activity Tracking Module\r\n\r\nReports & Visualization Module\r\n\r\nAdmin Management Module$$\n$$\nDietMate: A Mobile App for Monitoring Diet through Image-Based Food Recognition and Nutrition Analysis.\r\nGet Fit-Fitness Exercises & Diet Consultant$$\nAI-Based Food Recognition System$$\nRule-Based Personalized Diet and Workout Recommendation:$$\nEnhanced Allergy and Health-Aware Meal Suggestions:","comments":" $$ Ensure are all the changes incorporated in the revised proposal","isDraft":0,"status":2,"created_at":"2025-10-08 18:58:38","updated_at":"2025-10-27 14:36:12","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1046,"project_id":1417,"title":"RedLink: Blood Donor\u2013Receiver Matching Platform","prob":"In developing countries like Pakistan, currently there are no digital platforms that facilitate communication between blood donors and receivers. When needed, families of patients must manually search for the blood donors through social media groups, which is time-consuming. RedLink is a platform where receivers connect to the donors in real time. The existing systems are either manual, inefficient, or restricted and confined to a small group, which results to avoidable health risks. This project helps by offering a web-based solution that connects donors and receivers in real-time and makes sure that the donor\u2019s information is accurate and enables the users to request the blood by type and location.","description":"This system is a web-based blood donation platform developed using Python, Django and JavaScript for the backend and Vue.JS for the frontend. Two types of users will be using the system: the donors and the recipients. Both the donors and recipients will have to create their respective profiles first. The system will verify their login details. The donor profile will contain personal information, blood group, location, and contact details. Receivers can log in to place blood requests, including the type of blood needed as well as its location. Blood recipients can search the database for available donors based on blood type and location. This platform will also include an AI chatbot for blood mapping based on end-user's personal criteria. Web crawler will also be used to scrape blood donors' data from other websites.$$\nThe website\u2019s primary modules include (i) User Authentication and Authorization (ii) Profile Management Module (iii) Website Admin (iv) Blood Request Module (v) Donor Search Module (vi) Web Crawler for data scraping (vii) AI Chatbot$$\n(i) Profile Management Module (ii) Blood Request Module (iii) Implementing CRUD operations for donor\/receiver profiles such as blood group, location, and contact information (iv) REST APIs (v) AI chatbot.$$\n(i) Complete frontend interfaces (developed using Vue.js for responsive design) (ii) Website Admin (for content and user management) (iii) User Authentication and Authorization (for secure access control) (iv) Donor Search Module (to find donors by blood group and location) (v) Web crawler for data scraping.$$\n$$\n1. Blood Donation Website (donateblood.pk) \r\n2. Blood Donation Application (Save Life)$$\n1. AI chatbot.$$\n2. Web crawler for data scraping.$$\n3. Search donors by blood group and location.","comments":" $$ The revised project effectively addresses previous concerns by expanding scope and integrating AI and automation features. The concept is now technically sound and feasible, with meaningful real-world impact. Minor improvements include clearly defining the AI model\u2019s dataset, training process, and crawler data validation.","isDraft":0,"status":2,"created_at":"2025-11-04 00:02:51","updated_at":"2025-11-08 11:51:05","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1030,"project_id":1372,"title":"Heavy lift: Digital Market Place For Construction Equipment","prob":"Heavy machinery rentals are currently unorganized \r\nand inefficient. Contractors and farmers can\u2019t \r\nquickly find nearby equipment, check fair prices, or \r\nsee if it\u2019s available when they need it. Owners don\u2019t \r\nhave one place to list their machines, handle \r\nbookings, and build trust. That leads to delays and \r\nextra costs. Heavy Lift puts everything in one place \r\nlistings, availability, and easy booking so both sides \r\nsave time and avoid\u00a0confusion.","description":"Heavy Lift Marketplace is a web-based rental platform that connects equipment owners with renters seeking heavy machinery such as excavators, loaders, tractors, and other construction or industrial equipment. The system allows providers to list their machines with full details, including specifications, rental rates, availability calendar, delivery options, and multiple images. Renters can easily search and filter machinery by category, price, or location using an interactive map to find suitable options quickly and efficiently.The platform also includes a Driver Management Module, which lets providers add and manage their own trained drivers. Driver information such as license number, training institute, experience, and availability is entered by the provider and verified by the admin. When listing a machine, the provider can attach one or more verified drivers, and during booking, an available driver is automatically assigned to operate the equipment. This ensures that every rented machine is handled by qualified, approved personnel, reflecting real industry practice where owners supply trained operators with their machinery. To maintain equipment quality, the system integrates a Maintenance Management Module. This feature enables providers to record and monitor maintenance activities for their machinery, including last service date, mechanic details, and next scheduled service. Machines marked as \u201cUnder Maintenance\u201d are temporarily unavailable for booking until verified by the admin, ensuring that only safe, well-maintained equipment is provided to renters. Together, these features make Heavy Lift Marketplace a reliable, transparent, and user-friendly solution that simplifies the machinery rental process while ensuring safety, professionalism, and trust between equipment owners and renters.$$\n1. Authentication & Roles\r\nThis module manages secure login\/logout and user registration. Role-based access is provided for three main roles: Admin, Provider, and Renter. Admins manage users and listings, Providers list equipment, and Renters browse and book.\r\n2. Machinery Listings\r\nProviders can add machinery details such as name, category, rental price, specifications, images, delivery options, and location on a map. Renters can view complete specifications and multiple photos before booking.\r\n3. Search, Filters & Map\r\nRenters can search equipment using keywords, categories, rental rates, or location filters. An integrated map displays nearby available equipment, enhancing accessibility and ease of discovery.\r\n4. Booking & Availability\r\nThis module allows renters to request bookings for specific dates. It maintains an availability calendar to prevent overlaps and double bookings. Providers can accept or reject booking requests, while renters are notified of status updates.\r\n5. Messaging\r\nA built-in chat system enables direct communication between renters and providers. It supports sending inquiries, discussing terms, and negotiating before finalizing rentals.\r\n6. Payment\r\nA secure Stripe payment gateway is integrated into the system to handle online transactions. Renters can pay securely for bookings, while providers receive payment confirmations instantly.\r\n7. Notifications\r\nAutomatic notifications are triggered for all key actions such as booking confirmations, cancellations, and payment updates, keeping all users informed in real time.\r\n8. Reviews & Ratings\r\nAfter each completed rental, renters can provide ratings and reviews about the equipment and provider. This ensures transparency, helps renters make informed decisions, and builds trust.\r\n9. Admin Panel\r\nThe admin has full control over users, listings, drivers, payments, and system performance. Through the Analytics Dashboard, the admin can monitor renters, providers, categories, and transactions to ensure transparency and proper management.\r\n10. Analytics Dashboard\r\nAdmins and providers can view dashboards showing total users, bookings, payments, pending amounts, and rent-out history, along with daily, weekly, and monthly performance trends.\r\n11. Support\/Disputes\r\nThis module handles customer support and dispute resolution. Renters and providers can report issues, and admins can intervene to resolve conflicts fairly.\r\n12. Driver Module\r\nAllows machinery providers to add and manage their own drivers with details like license, training institute, experience, and availability. Admin verifies all information, and the provider assigns a verified driver to operate the booked machinery.\r\n13. Maintenance Management\r\nThis module lets providers record and track maintenance details such as service date, mechanic name, maintenance type, and next scheduled service. Machines marked \u201cUnder Maintenance\u201d are temporarily unavailable for booking until verified by the admin, ensuring renters always receive well-maintained and safe equipment.$$\n1. Search, Filters & Map: Allows renters to search machinery by category, price, and location using an integrated map for easy access.\r\n2. Messaging: Enables direct chat between renters and providers for quick inquiries and negotiations.\r\n3. Notifications: Sends automatic alerts for booking confirmations, cancellations, and payment updates in real time.\r\n4. Reviews & Ratings: Lets renters rate and review equipment and providers to build trust and transparency.\r\n5. Support\/Disputes: Handles complaints and conflicts with admin intervention for fair resolution.\r\n6. Driver Module: Allows providers to add and manage trained drivers with verified licenses and experience to operate booked machinery.\r\n7. Maintenance Management: Enables providers to record and track service or repair activities, marking machines under maintenance as unavailable until verified by admin to ensure safety and\u00a0reliability.$$\nI am building the following modules as part of my Final Year Project:\r\n1. Authentication & Roles: Manages secure login\/logout and registration with role-based access for Admin, Provider, and Renter.\r\n2. Machinery Listings: Allows providers to add machinery with details like name, category, price, specifications, location, images, delivery options, and attached driver details.\r\n3. Booking & Availability: Handles booking requests, maintains an availability calendar, and prevents overlapping or double bookings.\r\n4. Payment (Stripe Integration): Processes secure online payments through Stripe. Renters can pay safely, and providers receive instant confirmations.\r\n5. Analytics Dashboard: Provides admins and providers with real-time insights, showing total bookings, payments, pending amounts, and rent-out history with daily, weekly, and monthly performance trends.\r\n6. Admin Panel: Provides full control to the admin for managing users, listings, drivers, and payments, with access to system analytics for transparent monitoring.$$\n$$\n$$\n$$\n$$\n","comments":" $$ The revised proposal demonstrates clear technical improvement with added depth through maintenance, driver, and payment modules.","isDraft":0,"status":2,"created_at":"2025-10-17 11:18:56","updated_at":"2025-10-27 14:33:03","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1039,"project_id":1370,"title":"Multi-Branch Electronic Goods Inventory Management System","prob":"Managing inventory across multiple electronics retail branches is operationally challenging. Business owners struggle with maintaining accurate, real-time stock visibility across all locations. This leads to critical issues like stockouts of popular items in one branch while excess inventory sits in another, resulting in lost sales and tied-up capital. Manual tracking via spreadsheets causes data inconsistencies and prevents a unified view of business performance. The lack of automated low-stock alerts and predictive demand forecasting leads to delayed or inaccurate replenishment orders. Staff also face inefficient processes for inter-branch stock transfers. Furthermore, the absence of data-driven insights into product relationships misses strategic cross-selling opportunities, leaving revenue potential untapped. This complexity causes operational inefficiencies, reduces customer satisfaction, and hinders business growth.","description":"The Multi-Branch Inventory System is an intelligent web application that provides business owners with a central dashboard to manage inventory across all branches in real-time. It automatically updates stock levels with every sale, generates AI-powered alerts for demand forecasting and cross-selling opportunities, and creates purchase orders for suppliers. The system includes a Point-of-Sale (POS) interface for processing sales and printing receipts. Owners can view consolidated reports on product performance, compare branch profitability, and manage inter-branch stock transfers seamlessly. By replacing manual record-keeping, the system saves time, maximizes sales, reduces excess stock, and delivers predictive insights to drive business growth through an intuitive, user-friendly interface.$$\n1) Inventory Management Module: Tracks real-time stock levels, updates count automatically, provides low-stock alerts, manages branch transfers, and maintains inventory history. \r\n2) Product Catalog Module: Manages product information, specifications, categories, and pricing for both branches while maintaining separate databases. \r\n\r\n3) Purchase Order Module: Creates and tracks supplier orders, maintains vendor database, and generates automatic purchase suggestions.\r\n\r\n 4) Sales Order Module: Processes customer sales through POS interface, generates \r\nreceipts, manages returns, and updates inventory automatically.\r\n\r\n 5) Branch Management Module: Provides centralized control with role-based access for owners and branch managers and enables performance \r\ncomparison.\r\n\r\n 6) AI-Powered Business Intelligence Engine: A unified module that leverages machine learning on sales data to provide two key functionalities: \r\na) Demand Forecasting: Predicts future product demand for each branch by analyzing historical sales, seasonality, and trends. This enables proactive inventory planning, optimized \r\nstock levels, and automated purchase suggestions to prevent stock outs and reduce overstocking. \r\n\r\nb) Product Association & Cross-Selling: Identifies products frequently bought together and provides real-time suggestions to staff at the point of sale directly, increasing average order value and sales.\r\nc) Customer Feedback Analysis: The system analyzes customer feedback entered by staff to identify popular and low-rated products, helping owners decide which items to restock, improve, or replace.\r\n\r\nd) Smart Performance Insights: AI automatically checks which products and branches are performing best. It highlights slow-moving items and gives suggestions to improve branch performance and stock management.\r\n\r\n7) Notification Module: Sends automated alerts for low stock, sales targets, pending orders, and system updates.$$\nI will develop Inventory Management, Product Catalog, and Sales Order modules. His focus will be on creating user friendly interfaces for stock, products, and sales management. He will also handle the AI submodules for Product Association & Cross-Selling and Smart Performance Insights to improve sales and analyze branch performance.$$\nI will develop the Purchase Order, Branch Management, and Notification modules. He will manage automated purchasing, multi-branch operations, and real-time alerts. Additionally, he will work on the AI submodules for Demand Forecasting and Customer Feedback Analysis to predict stock needs and evaluate customer satisfaction.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Ensure all changes incorporated in the revised version.","isDraft":0,"status":2,"created_at":"2025-10-21 23:16:18","updated_at":"2025-10-31 10:29:23","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1044,"project_id":1330,"title":"StudyHub-An study consultancy Application","prob":"Many students who wish to pursue higher education abroad face challenges in:\r\n\u2022\tFinding reliable consultancy services,\r\n\u2022\tUnderstanding complex admission requirements,\r\n\u2022\tCorrectly applying to international universities recommended to student according to there Score, budgets and other things that related to student that student provide before he find the the university system will recommend student according to his\/her requirement. \r\n\u2022\tIdentifying suitable scholarships, and\r\n\u2022\tTracking their admission progress this will done the one Ai model review the application and marked it approved, denied etc. admin will just check the applications which he\/she thinks that i should check this application thoroughly otherwise AI model reviewed it already.\r\nStudent also have his consultancy through AI consultant chatbot or an separate consultant that is available. \r\nCurrent consultancy services are often costly, scattered, and not AI-powered. Students frequently miss deadlines, upload incorrect documents, or choose programs that don\u2019t match their academic background.","description":"StudyHub is a mobile-based application developed in Flutter with a Firebase backend. The application provides end-to-end consultancy and guidance services to students applying abroad for higher education.\r\nThe AI model suggesting suitable university will work on data like country , your budget, Any specific university \r\nand the criteria is merit your recent academic performance suggest you that this university is better you will more likely able to get admission in this university or predict that your chance of getting admission in this university is 70%. \r\nKey features:\r\n1. Student Module\r\n\r\nRegister\/login\r\n\r\nManage profiles\r\n\r\nSearch universities\/programs\r\n\r\nSubmit applications\r\n\r\nTrack status\r\n\r\n2. Admin Module\r\n\r\nManage universities & programs\r\n\r\nVerify documents\r\n\r\nMonitor applications\r\n\r\nGenerate reports\r\n\r\n3. Recommendation Engine (AI\/ML)\r\n\r\nSuggests suitable universities & programs based on student profile.\r\n\r\nUses content-based & collaborative filtering.\r\n\r\n4. Predictive Admission Model (ML)\r\n\r\nPredicts probability of admission using historical student data.\r\n\r\nExample: \u201cYou have 75% chance for University A.\u201d\r\n\r\n5. NLP Chatbot (AI Counselor)\r\n\r\nProvides 24\/7 automated guidance.\r\n\r\nAnswers FAQs on admissions, visas, scholarships.\r\n\r\n6. Scholarship & Financial Aid Module\r\n\r\nCentralized scholarship database.\r\n\r\nAI recommends scholarships based on academic\/financial background.\r\n\r\n7. Document OCR & Validation\r\n\r\nScans and verifies academic transcripts, certificates, and letters.\r\n\r\nUses OCR & AI to detect missing or invalid documents.\r\n\r\n8. Sentiment Analysis on Reviews\r\n\r\nStudents review consultancy services\/universities.\r\n\r\nAI analyzes tone (positive\/negative\/neutral).\r\n\r\n9. Payment & Agreement Module\r\n\r\nOnline fee payments for consultancy services.\r\n\r\nDigital agreements\/contracts.\r\n\r\n10. Analytics & Insights Module\r\n\r\nAdmin dashboard: applicants, success rates, popular universities.\r\n\r\nStudent dashboard: personalized insights (e.g., \u201cTop 5 programs that match your profile\u201d).\r\nafter application reviewed student will have an agreement and then payment process. and also one thing this project is for one consultant compony.$$\nThis system includes a Student Module for registration\/login, profile management, university\/program search, application submission, and status tracking. The Admin Module manages universities\/programs, verifies documents, monitors applications, and generates reports. An AI Recommendation Engine suggests universities and programs using content-based and collaborative filtering, while a Predictive Admission Model estimates admission chances based on historical data. An NLP Chatbot acts as a 24\/7 counselor, answering FAQs on admissions, visas, and scholarships. The Scholarship & Financial Aid Module provides a centralized database with AI-driven scholarship recommendations. OCR & Document Validation scans and verifies academic documents, while Sentiment Analysis evaluates reviews of universities and consultancy services. A Payment & Agreement Module supports online payments and digital contracts, and an Analytics & Insights Module offers dashboards with real-time statistics, trends, and personalized recommendations for students and admins.$$\nModule developed by SIDRA WANIA (Frontend & AI Integration)\r\n\u2022\tStudent Module (UI in Flutter), Scholarship & Recommendation UI, Chatbot integration, and Designing and implementing all frontend features.$$\nModule developed by MUHAMMAD FAIQ ADNAN (Backend & AI\/ML Models)\r\n\u2022\tAdmin Module (backend management), Predictive Admission Model, Document OCR & Validation, Sentiment Analysis, Payment & Agreement Integration, and Backend APIs for AI\/ML models.$$\n$$\n$$\n$$\n$$\n","comments":" $$ Ensure all changes incorporated in the revised version.","isDraft":0,"status":2,"created_at":"2025-10-27 19:00:58","updated_at":"2025-10-31 10:28:31","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}
No. Project Title Reg. No. Name Date & Time Evaluation Status
1. AI-Based Elderly Healthcare Assistant CIIT/FA22-BCS-025/WAH
CIIT/FA22-BCS-050/WAH
 MOAZ
MUHAMMAD AHMER
 Done Accepted  
2. Cloth Warehouse Inventory Management System CIIT/FA22-BSE-009/WAH
CIIT/FA22-BSE-021/WAH
 SARMAD ALI DANISH
MUHAMMAD ARHAM
 Done Accepted  
3. Insight AI: “AI-Powered Data Analytics for Smarter Decisions.” CIIT/FA22-BCS-018/WAH
CIIT/FA22-BCS-163/WAH
 MUHAMMAD JIBRAN YOUNAS
MUHAMMAD FAIZAN KHAN
 Done Accepted  
4. Time Explorer-An engaging app where children discover history through fun, colorful adventures. CIIT/FA22-BSE-025/WAH
CIIT/FA22-BSE-098/WAH
 USMAN AYAZ
ABDUL MOEED
 Done Accepted  
5. Seerah Timeline CIIT/FA22-BSE-010/WAH
CIIT/FA22-BSE-026/WAH
 HASSAN KALEEM
HASNAIN ALI
 Done Accepted  
6. VocalID: Intelligent Voice Authentication with Spoof & Liveness Detection CIIT/FA22-BCS-056/WAH
CIIT/FA22-BCS-157/WAH
 ABDUL BASIT ZAHID
MUHAMMAD FARRUKH ZIA
 Done Accepted  
7. Dynamic Reinforcement Learning Firewall with Adaptive Threat Detection CIIT/FA22-BCS-079/WAH
CIIT/FA22-BCS-081/WAH
 SHAYAN HAIDER
FAHAD HAROON
 Done Accepted  
8. EmoTales AI: Emotion-Aware Educational Story & Comic Generator CIIT/FA22-BCS-047/WAH
CIIT/FA22-BCS-126/WAH
 MUQADAS NAZ
ADEENA REHAN
 Done Accepted  
9. Cogniscan: AI-powered App for early dyslexia and dysgraphia detection CIIT/FA22-BCS-054/WAH
CIIT/FA22-BCS-098/WAH
 MUHAMMAD TALHA
JAWAD ZAHEER
 Done Accepted  
10. Manzil - Smart Career Counseling Portal CIIT/FA22-BCS-105/WAH
CIIT/FA22-BCS-147/WAH
 ABDUL BASIT ATTIQUE
TAHA IQBAL
 Done Accepted  
11. Drive Doctor: Smart Vehicle Maintenance Assistant CIIT/FA22-BSE-014/WAH
CIIT/FA22-BSE-032/WAH
 MUHAMMAD FARHAN AMEEN
IRTAZA KAZIM
 Done Accepted  
12. EvaluTech – AI-Powered Interview & Hiring Management Platform for “Factory Web Services” CIIT/FA22-BSE-048/WAH
CIIT/FA22-BSE-077/WAH
 AHAD UR REHMAN
RAHEEL AHMED
 Done Accepted  
13. HealthSync CIIT/FA22-BCS-038/WAH
CIIT/FA22-BCS-116/WAH
 ANWAAR AHMED
MUHAMMAD MUZAMMAL ARIF
 Done Accepted  
14. PMNH BioConnect CIIT/FA22-BCS-115/WAH
CIIT/FA22-BCS-158/WAH
 HASNAIN FIAZ
MUHAMMAD SAAD
 Done Accepted  
15. Zarkhez (Automated Irrigation Control) CIIT/FA22-BSE-015/WAH
CIIT/FA22-BSE-055/WAH
 SANA ULLAH
NAYAB ZAHRA
 Done Accepted  
16. Skill Swap CIIT/FA20-BSE-073/WAH
CIIT/FA22-BSE-111/WAH
 RANA MUHAMMAD ZULQARNAIN
MUZAMIL MUNIR
 Done Accepted  
17. Share4Good- A Smart Platform for Donors and Recipients CIIT/FA22-BSE-068/WAH
CIIT/FA22-BSE-078/WAH
 UMM-E-LILA
MARIA WAHID
 Done Accepted  
18. Road Damage Detection System CIIT/FA22-BSE-013/WAH
CIIT/FA22-BSE-118/WAH
 MOMINA MALIK
AQSA SHAHID
 Done Accepted  
19. CUI-Wah Job Fair Company and Admin Portal CIIT/FA22-BCS-073/WAH
CIIT/FA22-BCS-082/WAH
 SULIMANA HUMA
SHUMAIM ZAFAR
 Done Accepted  
20. Bridal Ease – A Smart Bridal Dress Renting and Selling Platform for Personalized Bridal Experience CIIT/FA22-BCS-058/WAH
CIIT/FA22-BCS-059/WAH
 SIDRA ILYAS
ALIZA MALIK
 Done Accepted  
21. AutoFix: Intelligent Multi-Branch Vehicle Service Hub CIIT/FA22-BCS-127/WAH
CIIT/FA22-BCS-144/WAH
 ABDUL WASAY
AHMAD ALI
 Done Accepted  
22. AI-Driven Smart Enterprise Resource Planning (ERP) System for TradeWell CIIT/FA22-BCS-001/WAH
CIIT/FA22-BCS-124/WAH
 MUHAMMAD BILAL TARIQ
UMER ALI
 Done Accepted  
23. CUI Wah Job Fair Student Portal CIIT/FA22-BCS-155/WAH
 MUHAMMAD HASSAN ASKARI
 Done Accepted  
24. Smart Food Waste Reduction and Donation App CIIT/FA22-BCS-096/WAH
CIIT/FA22-BSE-005/WAH
 TAFSEER-E-NOOR HASHMI
MANAHIL SHAHID
 Done Accepted  
25. Social Platform Exposure Tracker (SafeSocial) CIIT/FA22-BCS-165/WAH
CIIT/FA22-BCS-167/WAH
 ZARYAB AHMAD
SHAHZAIB HAIDER
 Done Accepted  
26. The Digital Scholar CIIT/FA22-BCS-071/WAH
CIIT/FA22-BCS-072/WAH
 MAAZ ABDULLAH
HANZALA SHAFIQUE
 Done Accepted  
27. Web Application Security Scanner CIIT/FA22-BCS-011/WAH
CIIT/FA22-BCS-122/WAH
 YASIR ALI KHAN WAZIR
GUL FARAZ KHAN
 Done Accepted  
28. HealthInsura360 CIIT/FA21-BCS-041/WAH
CIIT/FA22-BCS-007/WAH
 QURAT-UL-AIN AZHAR
KAYANAT MUGHAL
 Done Accepted  
29. DeepHunt AI – “Smarter search, Bigger Opportunities.” CIIT/FA22-BCS-003/WAH
CIIT/FA22-BCS-097/WAH
 ZUBAIR SHAFI
HASEEB AHMED
 Done Accepted  
30. Skinalytics: The Intelligent Path to Better Skin CIIT/FA22-BCS-085/WAH
CIIT/FA22-BCS-168/WAH
 AHMER MOON MAJID
MUHAMMAD HASSAN SHAHZAD
 Done Accepted  
31. FYP Review Management System (RMS) 5.0 CIIT/FA22-BSE-023/WAH
CIIT/FA22-BSE-031/WAH
 MUJTABA NAWAZ
ALI HASSAN
 Done Accepted  
32. Smart Meal Scanner, Nutrition Estimator & Weight Loss Planner CIIT/FA22-BCS-108/WAH
CIIT/FA22-BCS-136/WAH
 AYAAN ARSHAD
TAHA UMAR
 Done Accepted  
33. RedLink: Blood Donor–Receiver Matching Platform CIIT/FA22-BSE-064/WAH
CIIT/FA22-BSE-106/WAH
 HASHIR AHMED KHAN
MOHAMMAD TEHSEEN MURTAZA
 Done Accepted  
34. Farm2Market – AI-Powered Direct Farmer-to-Shopkeeper Platform CIIT/FA22-BSE-007/WAH
CIIT/FA22-BSE-027/WAH
 ALI HASSAN
SIDRA AKHTAR
 Done Accepted  
35. Heavy lift: Digital Market Place For Construction Equipment CIIT/FA22-BCS-121/WAH
CIIT/FA22-BCS-150/WAH
 AZEEMA SHAHEEN
MAHNOOR ASIF
 Done Accepted  
36. Multi-Branch Electronic Goods Inventory Management System CIIT/FA22-BSE-036/WAH
CIIT/FA22-BSE-039/WAH
 MUHAMMAD UMAIR NASEER
ARSALAN ALI
 Done Accepted  
37. StudyHub-An study consultancy Application CIIT/SP22-BSE-031/WAH
CIIT/SP22-BSE-032/WAH
 SIDRA WANIA
MUHAMMAD FAIQ ADNAN
 Done Accepted  
Committee: 3
Team Members:
Dr. Saima Gulzar Ahmed (HEAD)
Ms. Maha Rasheed
Ikram Ul Haq
Ms. Sania Umer
Venue: CS Conference Room 1
Remarks: Please reach the venue well in time, also prepare few slides to present your proposal, also bring your laptop

1{"id":955,"project_id":1390,"title":"RecoMart","prob":"Current e-commerce platforms face critical limitations that negatively affect the overall shopping experience. Product classification is often performed manually by store owners, resulting in frequent errors, inconsistencies, and scalability challenges. Search functionalities are typically limited, lacking the ability to interpret natural language queries or support visual search, leaving customers unable to find products efficiently through text or images. Recommendation systems are generally generic, providing poorly matched suggestions rather than personalized, context-aware recommendations. Furthermore, customer support is restricted to rule-based chatbots that cannot understand or respond to complex product-related queries. These shortcomings contribute to customer frustration, reduced engagement, lost sales opportunities, and inefficient platform management.\r\n\r\nThere is an urgent need for an intelligent, AI-powered e-commerce solution that can automatically classify products, enable natural language and visual search, deliver personalized recommendations, and provide advanced customer support through intelligent conversational agents.","description":"The AI-Powered E-commerce Platform is an advanced MERN stack web application designed to transform the online shopping experience through seamless integration of artificial intelligence. By combining modern web technologies with state-of-the-art AI models such as ResNet50, YOLOv8, and NLP-based systems, the platform delivers a smart, efficient, and user-centric shopping ecosystem.\r\nThe system is structured around three core interfaces:\r\nCustomer Portal \u2013 Enables users to browse, search, and purchase products with features like text-based and image-based search, personalized recommendations, and AI chatbot assistance.\r\nVendor Dashboard \u2013 Provides vendors with automated product classification using computer vision, inventory management, and sales analytics to optimize business operations.\r\nAdmin Panel \u2013 Supports administrators with tools for user management, order monitoring, and AI system performance tracking, ensuring smooth platform governance.\r\nKey features include:\r\nIntelligent Search \u2013 Natural Language Processing (NLP) enables customers to search products using everyday language.\r\nVisual Search \u2013 Users can upload images to find visually similar products through deep learning\u2013based models.\r\nSmart Recommendations \u2013 A recommendation engine analyzes user behavior to deliver personalized product suggestions.\r\nAI Chatbot \u2013 Understands and responds to queries, enhancing customer support beyond basic rule-based systems.\r\nVendor Support Tools \u2013 Automatic product categorization, inventory updates, and detailed sales insights.\r\nPlatform Management \u2013 Secure payment processing, real-time inventory synchronization, and full mobile responsiveness across devices.\r\n\r\nBy integrating AI across all layers of the system, the platform eliminates manual errors, enhances customer engagement, and streamlines vendor operations, ultimately creating a next-generation e-commerce experience.$$\n1. Customer Module: Handles user registration, authentication, product browsing, smart search functionality (text, image-based), shopping cart management, wishlist features, secure checkout process, order tracking, and review system. Integrates AI chatbot for customer queries and personalized product recommendations.\r\n2. Vendor Dashboard Module: Manages vendor registration and authentication, product upload with automatic AI categorization, inventory tracking, order processing, sales analytics, product performance monitoring, and bulk product management capabilities.\r\n3. Admin Panel Module: Provides comprehensive platform management including user and vendor oversight, product approval workflows, order monitoring, revenue analytics , AI model performance monitoring, and system configuration settings.\r\n4. AI Integration Module: Connects existing trained models (ResNet50 for image classification, YOLOv8 for object detection, NLP for text processing) through APIs, handles automatic product categorization, enables visual search functionality, powers recommendation engine, and supports chatbot natural language understanding.\r\n5. Payment Processing Module: Integrates secure payment gateways (Stripe\/Razorpay), handles transaction processing, manages refunds, generates payment reports, and ensures PCI compliance.\r\n6. Analytics Module: Collects user behavior data, tracks product performance, generates sales reports, monitors AI system accuracy, and provides business intelligence insights.$$\n1. Customer Module \u2192 User Experience & Engagement\r\n2. AI Integration Module \u2192 AI & Intelligent Services\r\n3. Payment Processing Module \u2192 Financial & Transaction Management\r\n4. Database module$$\n1. Vendor Dashboard Module \u2192 Seller & Inventory Management\r\n2. Admin Panel Module \u2192 Platform & Governance Management\r\n3. Analytics Module \u2192 Data & Business Intelligence\r\n4. Backend module$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-09-25 13:41:58","updated_at":"2025-10-16 10:12:43","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1001,"project_id":1413,"title":"eDevice Registration & Monitoring Portal (eDRMP)","prob":"In Pakistan, numerous mobile devices operate without proper registration, leading to theft and illegal resale. Existing systems like PTA\u2019s DIRBS provide registration and blocking, but they lack real-time theft prediction, police integration, and user awareness features.\r\neDRMP introduces an intelligent, mobile-based solution that not only registers and blocks devices but also uses AI and smart alerts to detect, prevent, and respond to theft activities.\r\nThe system includes separate admin panels for users, PTA officials, and police, integrates theft prediction alarms based on FIR location data, and provides real-time alerts to users near high-risk theft zones.","description":"The proposed system, eDRMP (Electronic Device Registration & Monitoring Portal), is a smart mobile-based application designed to register, track, and protect mobile devices from theft and misuse. Unlike the existing PTA DIRBS system, eDRMP introduces advanced functionalities such as a dedicated police admin panel for FIR management, AI-based theft prediction, and real-time theft alerts.\r\nUsers can register their devices using CNIC and IMEI, submit FIRs in case of theft, and track their device status. The police panel allows law enforcement to verify FIRs, manage stolen-device reports, and update case statuses. Using stored FIR location data, eDRMP predicts high-risk theft areas and notifies users through smart theft-protection alarms when they enter those zones.$$\nRegistration Module:\r\nThis module will allow users to register their mobile devices by entering CNIC and IMEI numbers. The data will be securely saved in the database for future verification and duplication prevention.\r\nFIR Module:\r\nUsers can file FIRs by entering FIR number, police station, and date, along with uploading FIR proof. The request will be forwarded to the Police Admin Panel for verification before any action is taken.\r\nBlocking Module:\r\nIn case of theft, users can request to block their devices. The request will only be processed after the FIR is verified by the police. Once approved, the device status will be updated as \u201cBlocked.\u201d\r\nTracking Module:\r\nThis module will help users and admin to track the real-time status of their device requests such as \u201cRegistered,\u201d \u201cPending,\u201d \u201cBlocked,\u201d or \u201cRejected.\u201d It also keeps record history for transparency.\r\nPolice Admin Panel:\r\nA dedicated interface for police officials to review and verify FIRs, manage theft reports, and mark cases as recovered or under investigation. It ensures faster coordination between police and PTA.\r\nTheft Prediction & Alarm Module:\r\nThis module uses stored FIR location data to identify theft-prone areas. When users enter such zones, the system will generate a smart theft alert notification to warn them in advance.$$\nRegistration Module:\r\nHandles device registration using CNIC and IMEI validation. Ensures secure data entry and prevents duplication.\r\nTracking Module: \r\nDisplays real-time updates for registration and blocking requests and maintains complete device activity history.\r\nPolice Admin Panel:\r\nProvides a separate dashboard for law enforcement to review and manage FIRs.$$\nFIR Module: \r\nManages FIR submission, uploading, and police verification process.\r\nBlocking Module: \r\nExecutes verified blocking requests and updates system records.\r\nTheft Prediction & Alarm Module: \r\nUses previous FIR location data to identify high-risk areas and generate real-time theft alerts for users.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-10-08 13:27:13","updated_at":"2025-10-20 09:56:56","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":989,"project_id":1410,"title":"FareSmart","prob":"In Pakistan, intercity bus passengers face difficulty when comparing fares, schedules, and seat availability across different service providers (e.g., Adil Coaches, Shaheen Express, Ziauddin Coaches). Each company manages its operations separately, making it hard for passengers to find the cheapest option, book seats, and track buses in real time. Our system solves this by providing a single mobile app and web platform where passengers can compare fares through interactive graphs, reserve seats (cash or online payment), view real-time seat availability, track buses, and receive alerts in both English and Urdu. This will improve transparency, save passenger time, and enhance trust in bus services.","description":"FareSmart is a professional mobile and web-based platform that integrates multiple intercity bus operators into one unified system. The application supports three actors: Admin, Driver, and Passenger.\r\n\r\nAdmin (Company\/Operator): Registers buses, drivers, routes, fares, and weekly schedules. Admin monitors bookings, reserved vs available seats, and earnings. A price comparison dashboard shows cheapest to most expensive buses across routes using interactive graphs.\r\n\r\nDriver: Registered by Admin and provided login credentials. Drivers can view their assigned trips, update trip status, share real-time location through GPS, and monitor passenger pickup stops.\r\n\r\nPassenger: Registers in the app, searches routes (e.g., Bannu \u2192 Islamabad), compares fares from different companies, selects pickup points, reserves seats (via cash or online payment), and receives notifications when the bus is near their stop. Passengers can also rate drivers and services for quality improvement.\r\nThe system uses Latitude\/Longitude for stops, calculates distance and estimated travel time, and provides real-time bus tracking. The platform is fully bilingual (English + Urdu).\r\n\r\nFareSmart is a cross-platform mobile and web application that brings together multiple bus operators under one platform. Admins manage buses, routes, drivers, fares, and schedules. Drivers use the mobile app to view trips, update status, and share real-time location. Passengers can search routes, compare prices using interactive graphs, reserve seats, select pickup points, and receive alerts when the bus is approaching. The system supports online and cash bookings, bilingual interface (English\/Urdu), and real-time seat availability. By combining all operators\u2019 information, FareSmart provides transparency, affordability, and convenience for intercity travelers.\r\n\r\nOverall, FareSmart solves the problem of fragmented bus booking systems by offering passengers a single, affordable, and intelligent platform with modern features like price comparison, seat visualization, booking history, notifications, and real-time tracking.$$\nAdmin Modules:\r\n1. Authentication & Profile Management\r\n2. Driver Registration & Management (CNIC & phone verification)\r\n3. Bus Registration & Management\r\n4. Route & Stop Management (Latitude\/Longitude, distance, duration)\r\n5. Fare & Weekly Schedule Management (checkboxes Mon\u2013Sun)\r\n6. Booking & Seat Monitoring (reserved vs available)\r\n7. Price Comparison Dashboard (interactive fare graphs)\r\n8. Reports & Analytics (income, seat utilization, ratings)\r\n\r\nDriver Modules:\r\n1. Authentication (Admin-provided credentials)\r\n2. Route & Schedule Viewing\r\n3. Trip Management (start\/end trip, update status)\r\n4. Live GPS Location Sharing\r\n5. Passenger List & Stop Notifications\r\n6. View Passenger Ratings\r\n\r\nPassenger Modules:\r\n1. Authentication & Profile (English\/Urdu toggle, CNIC verification)\r\n2. Route Search & Bus Comparison\r\n3. Seat Reservation (Cash \/ Online booking)\r\n4. Pickup Point Selection & Notifications (bus ETA, 5 km alert)\r\n5. Live Bus Tracking on Map\r\n6. Booking History & E-Tickets (QR code)\r\n7. Feedback & Ratings for drivers and bus companies$$\nIqra Noor (Mobile App \u2013 Passenger + Driver focus):\r\n1.\tPassenger Authentication & Profile (English\/Urdu, CNIC verification)\r\n2.\tRoute Search & Bus Comparison (with interactive graphs)\r\n3.\tSeat Reservation & Payment (Cash\/Online, QR ticket)\r\n4.\tPickup Point Selection & Notifications (bus ETA, 5 km alert)\r\n5.\tDriver App Features (Login, View schedule, Start\/End trip, Passenger list)$$\nSibgha Noor (Web + Admin Dashboard + Backend focus):\r\n1.\tAdmin Authentication & Profile Management\r\n2.\tDriver Registration & Management (by Admin)\r\n3.\tBus & Route Management (including Latitude\/Longitude stops, distance, duration)\r\n4.\tFare & Weekly Schedule Management (Mon\u2013Sun checkbox interface)\r\n5.\tBooking Monitoring + Analytics (reserved vs available seats, revenue, utilization, ratings)$$\n$$\nGeneric Online Bus Ticket Reservation System$$\nInteractive Price Comparison Dashboard across multiple bus operators.$$\nBilingual Support (English & Urdu) for inclusivity.$$\nReal-time GPS Tracking with Proximity Alerts for passengers.","comments":"","isDraft":1,"status":1,"created_at":"2025-10-07 18:30:18","updated_at":"2025-10-20 09:54:41","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1003,"project_id":1425,"title":"VisionMart","prob":"This Final Year Project aims to address key security and operational issues faced by retail marts. Shoplifting and suspicious customer actions cause major financial losses, while manual CCTV monitoring is inefficient and unreliable. At the same time, store managers often lack accurate data about customer inflow and shelf popularity, making it difficult to plan staffing, restocking, and promotions effectively.\r\n\r\nThe proposed system tackles these challenges through three modules:\r\n\r\nSuspicious Activity Detection \u2013 Automatically identifies theft attempts and abnormal behavior, reducing losses and improving security.\r\n\r\nPeople Counting \u2013Monitors entry and exit in real time to accurately identify peak shopping hours.\r\n\r\nCrowded Shelf Detection \u2013 Identifies shelves with high customer activity, providing insights into sales trends and helping increase product sales.\r\n\r\nTogether, these modules deliver a smart, automated surveillance solution that enhances security, supports business decisions, and improves overall customer experience.","description":"This Final Year Project proposes an AI-powered Smart Mart Surveillance System that addresses major security and operational challenges faced by retail marts. The system leverages computer vision techniques and real-time analytics to automate surveillance, reduce losses from theft, and provide valuable insights into customer behavior. Unlike traditional CCTV monitoring, which is manual and error-prone, this system delivers intelligent, data-driven solutions that improve both security and business operations.\r\n\r\nModule 1: Suspicious Activity Detection\r\n\r\nThis module continuously monitors customer actions through surveillance cameras. Using YOLO-based object detection and action recognition models, the system automatically identifies abnormal or theft-related behaviors, generating instant alerts with supporting video evidence. This helps reduce financial losses and lessens reliance on manual supervision.\r\n\r\nModule 2: People Counting\r\n\r\nThis module tracks the number of customers entering and exiting the mart in real time. Through line-crossing logic applied to object detection, it provides accurate inflow and outflow statistics. More importantly, it identifies peak shopping hours, enabling managers to plan promotional campaigns during high-traffic times and thereby maximize sales opportunities.\r\n\r\nModule 3: Crowded Shelf Detection\r\n\r\nThis module highlights shelves that attract the most customer activity. By analyzing crowd density, the system provides insights into sales trends and helps identify which products generate higher demand. This assists store managers in planning stock levels and developing sales-focused strategies.\r\n\r\nIntegration and Conclusion\r\n\r\nAll processed data and alerts are delivered to a Flutter-Firebase mobile application, ensuring real-time accessibility for mart managers. The app presents suspicious activity alerts, entry\/exit statistics, and insights into crowded shelves, thereby combining security and business intelligence in a single platform.\r\n\r\nIn summary, this project provides a comprehensive yet focused solution that enhances retail security, identifies peak business hours, and reveals sales-driven shelf activity\u2014ultimately improving both operational efficiency and profitability.$$\n1. Suspicious Activity Detection\r\nRetail marts frequently face theft and abnormal customer actions, resulting in significant financial losses. Conventional surveillance systems rely on manual CCTV monitoring, which is time-consuming and prone to human error. To address these issues, this module introduces an automated suspicious activity detection system powered by computer vision and deep learning. Live video streams from surveillance cameras are processed through YOLOv8 for object and person detection, combined with pose estimation and temporal behavior modeling to analyze customer movements and interactions. The system detects various suspicious actions, including putting items into personal bags or pocket, stealing from someone\u2019s pocket or bag, hiding objects under clothing, loitering for extended periods, weapon presence, and fighting behavior. Upon detection, the system generates real-time alerts with bounding boxes and video snippets, transmitted via Firebase to a Flutter-based mobile app for immediate staff response. This module enhances security, minimizes manual monitoring, and effectively reduces financial losses.\r\n\r\n2. People Counting (Entry\/Exit Tracking)\r\n\r\nAccurate monitoring of customer inflow and outflow is critical for effective mart management. The People Counting module fulfills this by using cameras at entry and exit points. It applies YOLO-based person detection together with a virtual line-crossing algorithm to count how many customers enter and leave the mart in real time.\r\n\r\nResults are continuously updated and displayed on the mobile application dashboard. A major advantage of this module is its ability to identify peak shopping hours by analyzing entry patterns over time. This insight is highly valuable, as managers can strategically launch targeted sales promotions during high-traffic periods to maximize revenue. Additionally, peak-hour analysis can support more effective resource allocation and staff scheduling.\r\n\r\nBy converting simple entry\/exit data into actionable insights, this module contributes to business growth and operational planning, helping marts optimize their performance.\r\n\r\n3. Crowded Shelf Detection\r\n\r\nA major challenge in marts is determining which shelves receive the most customer attention. Traditionally, this relies on staff observations, which are often inaccurate. The Crowded Shelf Detection module resolves this by using advanced crowd counting models combined with YOLO for person detection, to analyze customer density around shelves.\r\n\r\nThe system generates heatmaps and crowd density reports that highlight shelves frequently visited by customers. These outputs give managers clear insights into sales trends by showing which products and categories attract higher interest. Such knowledge enables optimization of product placement, timely restocking, and targeted promotional strategies.\r\n\r\nFor example, if a shelf consistently shows high customer activity, managers can prioritize stock management and even apply promotional offers to further boost sales. Thus, this module converts customer activity into sales-oriented intelligence, improving both marketing and inventory strategies.\r\n\r\nConclusion\r\n\r\nTogether, these three modules\u2014Suspicious Activity Detection, People Counting, and Crowded Shelf Detection\u2014form the foundation of the Smart Mart Surveillance System. The first strengthens security, the second highlights peak-hour sales opportunities, and the third reveals sales-driven shelf activity. Collectively, they deliver a comprehensive, AI-powered solution that integrates surveillance, business insights, and operational optimization for modern retail environments.$$\n1. AHAD ALI (CIIT\/FA22-BCS-143\/WAH) \u2013 Flutter Developer\r\nAhad Ali will be responsible for developing the mobile application using Flutter and integrating it with Firebase. His primary role is to design the frontend interface and ensure smooth visualization of the outputs generated by the computer vision modules. The modules assigned to him include:\r\nIntegration of Suspicious Activity Detection results into the app dashboard with real-time alerts.\r\nIntegration of People Counting data with visual charts and notifications, including peak-hour insights.\r\nIntegration of Crowded Shelf Detection outputs such as heatmaps and sales-related analytics.\r\nAlthough the modules are divided between Ahad Ali and Muhammad Nouman Afzal, the work will be carried out collaboratively by both members, covering both the application and computer vision aspects.$$\nMuhammad Nouman Afzal will be responsible for implementing the computer vision modules and backend processing. His role involves training and deploying models, extracting insights, and sending the processed results to Firebase for app integration. The modules assigned to him include:\r\nSuspicious Activity Detection \u2013 Implementing YOLOv8 with action recognition for identifying theft or abnormal behavior.\r\nPeople Counting \u2013 Using YOLO with line-crossing logic to calculate entry and exit counts, and deriving peak-hour insights.\r\nCrowded Shelf Detection \u2013 Applying crowd counting models with YOLO to identify congested shelves and generate heatmaps.\r\nAlthough the modules are divided between Ahad Ali and Muhammad Nouman Afzal, the work will be carried out collaboratively by both members, covering both the application and computer vision aspects.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-10-08 14:23:54","updated_at":"2025-10-16 10:07:13","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1041,"project_id":1406,"title":"Echo Lens (Industrial Project)","prob":"Telecom and enterprise brands like Nayatel face challenges in tracking customer sentiment and public perception across fast-moving social media platforms. Manual monitoring is inefficient, biased, and often too slow to detect service issues, customer dissatisfaction, or campaign performance trends. Echo Lens solves this by automatically collecting brand-related posts and comments from multiple online sources, analyzing their sentiment (positive, neutral, or negative), emotions, and discussion topics, and visualizing actionable insights on a web dashboard.\r\nIt also allows competitive benchmarking between two brands (e.g., Nayatel vs. PTCL) to identify market position, user satisfaction trends, and recurring service complaints. The system reduces manual workload, enhances monitoring coverage, and helps organizations make timely, data-driven decisions to improve customer experience and brand reputation.","description":"Echo Lens is an AI-powered web application designed to monitor brand sentiment and customer perception for organizations like Nayatel. Users can register a brand and optionally specify a competitor (e.g., Nayatel vs. PTCL). The system periodically collects relevant social media content via official APIs (e.g., Twitter\/X, Reddit) . The data is cleaned, filtered, and stored in a relational SQLite database.\r\n\r\nAn integrated NLP pipeline then performs language detection, entity recognition, sentiment analysis, emotion tagging, and aspect-based topic extraction using transformer-based models. It also generates concise textual summaries to explain the reasons behind customer sentiment shifts. The analytics module computes KPIs such as sentiment ratio, share of voice, trending topics, and complaint frequency.\r\n\r\nThe frontend dashboard allows users to visualize these results through interactive charts, keyword clouds, and comparative views between two brands. It supports filtering by time, platform, and sentiment category. Users can export reports (PDF\/CSV) and set alert thresholds for sudden negative sentiment spikes. Authentication and secure API key management ensure data safety.\r\n\r\nBy combining official APIs and controlled scraping, Echo Lens offers an industrial-grade solution for social sentiment tracking, enabling real-time brand intelligence and actionable insights for marketing, customer support, and management teams.$$\n1. Data Sources & API Connectors\r\nConnectors for X (formerly Twitter) and Reddit via official APIs. Handles auth, pagination, rate limits, retries, de-duplication, and input keyword lists per brand.\r\n\r\n2. Ingestion & Scheduler (Celery\/Beat)\r\nPeriodic jobs fetch new posts, validate schema, and persist to storage. Maintains fetch logs and back-off on API errors. Supports per-brand cadence and replays for missed windows.\r\n\r\n3. Preprocessing & Normalization\r\nLanguage detection, text cleaning (URLs\/mentions\/emojis), tokenization, brand and competitor entity matching, spam\/low-quality heuristics, and safe handling of user-generated content.\r\n\r\n4. NLP\/AI Pipeline\r\nTransformer-based sentiment (positive\/neutral\/negative), optional emotion tags, aspect\/topic keywords, and LLM-assisted summaries (\u201cWhat users like\/dislike today\u201d). Batch and on-demand inference modes with caching.\r\n\r\n5. Analytics & KPIs Engine\r\nComputes time-series sentiment, share of voice, top themes, influencer\/author stats (if available), and post-level scores. Supports A\/B comparison (two brands) with aligned time windows.\r\n\r\n6. Storage Layer\r\nSQLite collections for raw posts, processed annotations, aggregates, and audit logs. Indexes for keywords, brand IDs, timestamps, and sentiment to keep queries fast.\r\n\r\n7. Admin & Configuration\r\nBrand\/project setup, keyword\/hashtag lists, API keys\/quotas, alert thresholds, user roles (admin\/analyst\/viewer). Audit trails for changes.\r\n\r\n8. Web Dashboard (Frontend)\r\nReact.js interface with filters (date, platform, brand), charts (sentiment split, trends, word clouds), tables of top posts, and comparison views (Brand A vs Brand B). Drill-downs to original posts via links.\r\n\r\n9. Exports & Reporting\r\nCSV data exports and auto-generated PDF snapshots (with charts and headline insights) for weekly\/monthly reviews.$$\nI will develop the core backend, data processing, AI, and data layer. This includes API connectors (X\/Reddit), Celery-based schedulers, and preprocessing (language detection, cleaning, brand\/competitor recognition, spam filtering). I will integrate transformer models for sentiment\/emotions, implement aspect\/topic extraction, caching, and the aggregation engine for KPIs (trends, share of voice, alerts). I will design the SQLite schema, manage migrations, FTS5 indexing for fast search, query optimization, and backup procedures.$$\nI will focus on frontend and user interaction modules. This includes building the React.js web dashboard with visualizations (charts, sentiment timelines, comparison graphs, and word clouds). I will implement filtering options (date, platform, brand) and enable drill-downs to view specific posts. I will also create the Admin UI for adding brands, managing keywords, and handling user roles.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-22 15:22:16","updated_at":"2025-11-25 10:01:24","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":969,"project_id":1341,"title":"Orbi (Orbiting Your Life)","prob":"This project addresses the limitations of existing productivity tools like calendars and reminder apps, which lack human-like interaction and adaptability. Current systems provide basic notifications but fail to engage users in meaningful conversation, \r\nunderstand their emotional state, or proactively assist with daily tasks. Our virtual assistant introduces a realistic human avatar and an AI-driven personality that can adapt, respond naturally, and help users manage tasks in a conversational and interactive way.","description":"The Virtual Robot Assistant is an application developed using Flutter and integrates AI for natural language conversation, offers voice interaction, and features a realistic human avatar to enhance user engagement. The assistant manages daily tasks, provides proactive reminders, and adapts its personality based on user mood and habits. It combines modern conversational AI with productivity tools to deliver a personalized and intelligent user experience$$\nChat Module:\r\n\u2022\tThis module enables real-time text-based communication between the user and the assistant.\r\n\u2022\tIt uses the OpenAI API to understand user queries and generate natural, human-like responses.\r\n\u2022\tIt also maintains the context of the conversation to support multi-turn dialogues.\r\nVoice Interaction Module:\r\n\u2022\tProvides speech-to-text functionality so the assistant can listen to user commands.\r\n\u2022\tIntegrates text-to-speech so the assistant can reply in a natural human-like voice.\r\n\u2022\tEnsures smooth hands-free interaction for convenience and accessibility.\r\n1.\tAvatar Module\r\n\u2022\tDisplays a realistic animated human avatar that reacts during conversation.\r\n\u2022\tAvatar shows expressions like smiling, blinking, or nodding, giving the assistant a human-like presence.\r\n\u2022\tHelps create an engaging, friendly, and interactive user experience.\r\n2.\tTask Management Module\r\n\u2022\tAllows users to add, edit, delete, and track daily tasks.\r\n\u2022\tProvides reminders and notifications for upcoming tasks.\r\n\u2022\tKeeps a log of completed and pending tasks to help users stay organized.\r\n3.\tPersonality Engine\r\n\u2022\tDefines and manages the assistant\u2019s personality and behavior.\r\n\u2022\tAdapts tone and responses depending on user mood, context, and past interactions.\r\n\u2022\tMakes the assistant feel unique, engaging, and personalized compared to generic bots.\r\n4.\tMemory Module\r\n\u2022\tStores user\u2019s preferences, routines, and past interactions in Database.\r\n\u2022\tEnsures the assistant can remember past tasks and adapt to user habits.\r\n\u2022\tProvides context-aware responses (e.g., reminding of unfinished work or recurring routines).$$\nChat Module \u2192 She will connect the app with the GPT API so the assistant can understand text queries and reply like a human.\r\nVoice Interaction Module \u2192 She will add speech-to-text (to listen) and text-to-speech (to speak back), making the assistant voice-enabled.\r\nTask Management Module \u2192 She will create the system to add, edit, delete, and track tasks with reminders and notifications.\r\nDatabase Integration \u2192 She will manage data storage (user tasks, history, preferences) on Database for persistence.$$\nAvatar Module \u2192 Develop a realistic animated avatar (Lottie\/PNG\/3D) with expressions that react to user interaction.\r\nPersonality Engine \u2192 Implement the assistant\u2019s dynamic personality (tone, style, mood adaptation based on context and user input).\r\nUI\/UX Design \u2192 Design the overall interface (chat screen, avatar screen, task dashboard)$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-01 12:26:05","updated_at":"2025-10-15 11:01:53","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":958,"project_id":1331,"title":"SaamayAI","prob":"Quran recitation learners often struggle with mistakes such as skipping verses, missing words, or repeating sections, especially in the absence of a qualified teacher. Traditional supervision is limited, or sometimes inaccessible for millions of learners worldwide. As a result, learners may unknowingly develop persistent mistakes, which weakens their memorization (Hifz) and overall recitation confidence. SaamayAI solves this by providing an AI-powered personal tutor in the form of a mobile app.","description":"SaamayAI works on Saamay Speech llm which is a fine-tuned AI model designed specifically for Quran recitation. Unlike generic speech recognition, SaamayAI Model has been trained on Quranic audio-text pairs to ensure high accuracy for religious text. The user records their recitation through the Flutter app, which processes the audio and sends it to the SaamayAI model hosted via Hugging Face API. SaamayAI generates a transcription aligned with the Quranic text and identifies skipped, repeated, or misplaced verses. The app highlights mistakes.$$\n1) Recording Module: For recording user audio\r\n2) Audio Cleaning: Noise Suppression and downsampling module\r\n3) User Module: Includes profile creation, progress tracking\r\n4) Transcription Module: Transcribes user recorded audio to Arabic text\r\n5) SaamayAI Speech Recognition Module: Fine-tuned Whisper-based model specifically adapted for Quran recitation. Custom dataset of Quranic recitations ensures domain-specific accuracy.\r\n6) Live Audio Comparison Module: Compares transcription of live user audio against the Quranic verse database.\r\n7) Mistake Highlight Module: Highlights mistakes visually (missing words, wrong sequence).$$\n1) Recording Module \r\nFor recording user audio\r\n2) Audio Cleaning\r\nNoise Suppression and downsampling module\r\n3) User Module \r\nIncludes profile creation, progress tracking\r\n4) Transcription Module\r\nTranscribes user recorded audio to Arabic text$$\n1) SaamayAI Speech Recognition Module\r\nA fine-tuned Whisper-based model specifically adapted for Quran recitation. Custom dataset of Quranic recitations ensures domain-specific accuracy.\r\n2) Live Audio Comparison Module\r\nCompares transcription of live user audio against the Quranic verse database.\r\n3) Mistake Highlight Module\r\nHighlights mistakes visually (missing words, wrong sequence).$$\n$$\nAl-Qur\u2019an App: An AI-Powered Interactive Platform for Comprehensive Qur\u2019anic Learning$$\nFine-tuned Quran-Specific AI (SaamayAI) \u2013 Whisper is adapted and optimized for Quranic recitation, unlike generic ASR models.$$\nLive Comparison \u2013 Provides live audio comparison with original text$$\nAudio Preprocessing \u2013 Cleaning and optimizing audio to ensure clarity and accuracy during practice and evaluation.","comments":"","isDraft":0,"status":2,"created_at":"2025-09-29 13:59:22","updated_at":"2025-10-15 10:54:47","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":967,"project_id":1384,"title":"SmartStock: Conversational Inventory Intelligence System","prob":"Traditional inventory systems are complex, requiring SQL skills and IT support. This keeps \r\nvaluable business data locked away from owners, who must rely on technical middlemen to \r\ninterpret reports. Small and medium businesses then struggle to get timely, accurate insights, making it harder to avoid stock shortages or manage money effectively. Many are left with slow, error-prone manual reporting. SmartStock solves this problem with a powerful cloud LLM. Through a simple system, users can type or speak questions in plain English or Roman Urdu (for example: \"Aj ka profit kitna howa?\"). The system safely translates these questions into SQL using a secure medium, giving business owners direct, conversational access to their data. With SmartStock, insights that once required experts and delays are now available instantly helping businesses stay in control, prevent losses, and make smarter decisions with confidence.","description":"SmartStock is an advanced inventory intelligence system designed to help Small and Medium sized Enterprises (SMEs) manage their business data effortlessly. Many traditional inventory systems are complex, require SQL knowledge, and depend heavily on IT staff. This often leaves crucial data inaccessible to business owners and non-technical staff. SmartStock solves this problem by providing a simple, conversational interface, giving instant access to accurate insights while ensuring strong security. Users can type or speak queries in natural language, including English or Roman Urdu (for example: \u201cAj ka profit kitna howa?\u201d). The query is first processed by a critical logic layer that classifies requests as Query, Update, or Invalid, preventing unsafe or harmful operations. Once validated, the request, along with the database schema, is sent to the LLM. LLM generates precise SQL commands based on the user\u2019s intent. These commands are executed under strict Role-Based Access Control, ensuring security and controlled access to sensitive data. Finally, the raw output is converted into user-friendly formats. By combining natural language processing with secure, intelligent database handling, SmartStock removes technical barriers and empowers business owners and staff to access actionable insights directly. This enables them to make informed decisions, prevent stockouts, manage cash flow, and optimize operations in real time.$$\nIn SmartStock, the modules are,\r\n(i) User Interface (UI) Module \u2013 Lets users type or speak queries in English or Roman Urdu. \r\n(ii) Input Validation Module \u2013 Checks and cleans user input before processing. \r\n(iii) Intent Module \u2013 Classifies requests as Query, Update, or Invalid to prevent unsafe \r\nactions. \r\n(iv) LLM Query Generator \u2013 Converts validated queries into accurate SQL commands. \r\n(v) Database Module \u2013 Executes SQL queries securely. \r\n(vi) Response Module \u2013 Formats query results into tables, charts, or reports. \r\n(vii) Authentication & Authorization Module \u2013 Verifies user identity and manages role-based \r\naccess. \r\n(viii) Analytics Module \u2013 Provides insights like trends, forecasts, and KPI summaries. \r\n(ix) Alert & Notification Module \u2013 Sends warnings for stockouts, cash flow issues, or \r\nanomalies. \r\n(x) Reporting & Export Module \u2013 Allows exporting or sharing of visual reports and \r\ndashboards.$$\nIn SmartStock, I shall develop \r\n(i) User Interface (UI) Module \u2013 Lets users type or speak queries in English or Roman Urdu. \r\n(ii) Input Validation Module \u2013 Checks and cleans user input before processing. \r\n(iii) Intent Module \u2013 Classifies requests as Query, Update, or Invalid to prevent unsafe \r\nactions. \r\n(iv) LLM Query Generator \u2013 Converts validated queries into accurate SQL commands. \r\n(v) Database Module \u2013 Executes SQL queries securely.$$\nIn SmartStock, an intelligent system, I shall develop \r\n(i) Response Module \u2013 Formats query results into tables, charts, or reports. \r\n(ii) Authentication & Authorization Module \u2013 Verifies user identity and manages role-based \r\naccess. \r\n(iii) Analytics Module\u2013 Provides insights like trends, forecasts, and KPI summaries. \r\n(iv) Alert & Notification Module \u2013 Sends warnings for stockouts, cash flow issues, or \r\nanomalies. \r\n(v) Reporting & Export Module\u2013 Allows exporting or sharing of visual reports and \r\ndashboards.$$\n$$\n$$\n$$\n$$\nN","comments":"","isDraft":0,"status":2,"created_at":"2025-10-01 01:22:17","updated_at":"2025-10-15 11:46:08","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":994,"project_id":1411,"title":"ArtHive","prob":"Many talented small-scale artists struggle to gain visibility and recognition due to the lack of accessible digital platforms dedicated to art. Most of them neither have the resources to build personal websites nor the technical knowledge to manage online sales. As a result, their creativity remains limited to local communities, preventing them from reaching wider audiences and potential buyers. This project addresses that problem by providing a centralized online art gallery where artists can easily create accounts, upload their artworks, and showcase their talent to the public. Buyers, in turn, get a convenient platform to explore diverse art collections and purchase directly from artists. While other online marketplaces exist, this platform is specifically designed for art, making it more user-friendly and tailored to the unique needs of artists and art enthusiasts. Ultimately, it empowers creators and builds a focused, accessible marketplace dedicated solely to art","description":"Our project is a web-based multi-user art gallery platform designed exclusively for artists and art enthusiasts. The system enables small-scale artists to create accounts, upload their artworks, and showcase their talent to a wider audience. Each artist functions as a seller by posting art pieces with details such as title, description, price, and medium. Buyers act as users who can browse, search, and view artworks, add them to cart or wishlist, and directly purchase items of interest.\r\nWe used a CNN-based image classification Module trained on dataset\r\n(https:\/\/www.kaggle.com\/datasets\/doctorstrange420\/real-and-fake-ai-generated-art-images-dataset) to distinguish between human-made and AI-generated artworks by recognizing visual patterns such as brush-stroke irregularities, natural textures, and repetitive AI artifacts and also recommend price by analyzing attributes like size, medium, style, and historical prices artworks(https:\/\/www.kaggle.com\/datasets\/jijagallery\/art-market-dataset-selling-paintings-prediction)\r\nAdministrators manage the overall system by monitoring users, validating content, and ensuring security. This system bridges the gap between creative artists and potential customers.$$\nThe proposed system consists of four main modules: Artist Module, Buyer Module, Admin Module, and AI Module. \r\nThe Artist Module enables artists to register, log in, and manage their profiles. They can upload artworks with details such as title, description, price, and medium, while also editing or removing their listings. Artists are able to track sales, handle order requests, and interact with buyers.\r\nThe Buyer Module allows users to create accounts, browse and search artworks using filters like category, artist, medium, or price, and view detailed pages of each artwork. Buyers can add items to cart or wishlist, place orders, and review their purchase history. \r\nThe Admin Module is responsible for managing and verifying artist accounts, approving or removing artworks to maintain quality, and monitoring buyer accounts. Admins also handle disputes, generate analytical reports on sales and traffic, and ensure system performance, security, and database integrity. \r\nIn AI module, We will use a CNN-based image classification model trained on the Kaggle dataset (https:\/\/www.kaggle.com\/datasets\/doctorstrange420\/real-and-fake-ai-generated-art-images-dataset).This dataset contains two labeled classes real human made artworks and AI generated artworks which allows the CNN to learn distinctive visual patterns such as brush stroke irregularities, natural texture, and noise in real art versus smoother textures, repetitive patterns, and unnatural edges in AI-generated images.\r\nOnce trained, our model will analyze the uploaded artwork\u2019s pixel features and output a probability score indicating whether it is AI-generated or human-made, helping the admin decide on its authenticity before approval.\r\nThen for price recommendation we will use a CNN-based regression model trained on the dataset from Kaggle (https:\/\/www.kaggle.com\/datasets\/jijagallery\/art-market-dataset-selling-paintings-prediction). This dataset contains paintings with features such as medium, subject of painting(wild life, landscape) size (dimensions), style, and historical sale prices, among others. By feeding these attributes (and possibly image features) into the model, it will learn how each factor influences price (for example, large oil paintings by established artists tend to sell for more). When a new artwork is uploaded, our model will use its attributes size, medium, style, etc to output a recommended price range to assist the artist in pricing their work reasonably.$$\nIn this project, I will develop the Admin and Buyer modules, which include verifying sellers and artworks, monitoring users, resolving disputes, generating reports, along with buyer registration and login, browsing and searching artworks, viewing details, adding to cart or wishlist, placing orders, and checking purchase history.$$\nIn this project, I will develop the AI and Seller modules, which includes seller registration and login, profile management, uploading artworks with details like title, price and medium, updating portfolio, and tracking sales and orders.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-10-07 22:57:42","updated_at":"2025-10-16 11:30:03","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":992,"project_id":1396,"title":"InterStock-Stock Market Learning\u00a0Platform","prob":"Students studying finance and business lack practical exposure to stock market trading.\r\nLearning is mostly limited to theory, with no real-time practice opportunities.\r\nDirect entry into financial markets is risky for beginners due to possible monetary loss.\r\nTeachers have no effective tools to track and assess students\u2019 applied trading skills.\r\nAs a result, students graduate without confidence in handling real market scenarios.Our FYP, InterStock, solves this problem by providing a virtual stock trading platform that uses live market prices but fake money. This allows students to practice trading in a risk-free environment while gaining hands-on exposure. The platform includes leaderboards, assignments, and chat features to make learning interactive and engaging. Teachers benefit from a web dashboard to monitor progress & upload study material.. By bridging the gap between theory and practice, InterStock equips students with financial literacy, decision-making skills, and real-world confidence before entering the actual stock market.","description":"InterStock is a web-based learning platform that enables students to practice stock trading in a safe, risk-free environment. Students log in through the Student Dashboard, where they can trade stocks, options, and futures using live data with virtual money. They receive assignments, access learning material, and see performance on leaderboards.\r\nThe Teacher Dashboard allows instructors to manage students, upload content, generate reports, and organize trading competitions.\r\nThe system also includes AI and ML modules to evaluate student performance and predict potential profit\/loss based on trading decisions.\r\nThrough interactive dashboards, graphs, and smart assessment models, InterStock transforms theoretical finance education into real-world, hands-on learning.$$\nThe proposed system, InterStock, consists of multiple modules that work together to create an interactive and intelligent stock market learning environment. \r\n1.The User Authentication Module manages account creation, login, and verification for both students and teachers, ensuring secure access through role-based authentication and data encryption. \r\n2. The Student Dashboard Module serves as the main interface for learners, providing access to all essential features such as assignments, quizzes, chatrooms, study materials, notes, competitions, and teacher contact. It includes a Learn Screen for trading with virtual money using live stock prices, a Rank Screen displaying leaderboards, and a My Profile section for managing account settings and privacy controls.\r\n3. The Teacher Dashboard Module enables instructors to upload and track assignments or quizzes, send announcements, organize trading competitions, and monitor student progress and chatroom activity to maintain a safe learning environment.\r\n4. The Trading Simulator Module, which allows students to buy and sell stocks using live market data from APIs like Yahoo Finance and Alpha Vantage, helping them understand portfolio management, profit and loss calculation, and decision-making without financial risk.\r\n5. An intelligent AI & Analytics Module enhances the system\u2019s effectiveness by combining three key components: an AI assessment feature that evaluates students\u2019 performance by comparing quizzes, assignments, and trading outcomes; a Machine Learning model trained on historical stock data (using algorithms like Linear Regression or LSTM) to predict market trends and assess trading accuracy; and a Graph and Data Visualization component that presents student progress, portfolio growth, and profit\/loss statistics through interactive charts and graphs. \r\n6. The Assignments and Quizzes Module allows teachers to create, assign, and grade academic tasks while students receive timely notifications about submissions and deadlines. 7. The Leaderboard and Competition Module motivates learners through digital badges and rankings at class, school, and global levels, promoting healthy competition.\r\n8. The Communication and Chat Module supports group discussions and private messages in a monitored environment to ensure safety and discipline. \r\n9. The FAQ and Help Module offers instant assistance through a built-in chatbot that explains financial concepts, trading terms, and platform navigation, providing continuous support to users.$$\nIshmal Fatima (FA22-BCS-041) \u2013 Student Dashboard Module\r\nIshmal Fatima will develop the Student Dashboard Module, which provides students with access to assignments, quizzes, chatrooms, study materials, and competitions. She will also implement the Learn Screen for trading with virtual money, the Rank Screen for leaderboards, and My Profile for account management. This module ensures students gain interactive and practical trading experience in a safe environment.$$\nJawad-ur-Rahman (FA22-BCS-159) \u2013 Teacher Dashboard Module\r\nJawad-ur-Rahman will design the Teacher Dashboard Module, enabling teachers to manage students, upload study content, and send notifications about assignments and quizzes. He will also handle leaderboard tracking, generate reports, organize class competitions, and integrate chat monitoring. This module ensures teachers provide structured guidance while effectively evaluating student progress.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-10-07 22:14:47","updated_at":"2025-10-16 10:48:31","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1014,"project_id":1428,"title":"EduVest: An Equity-Funding Platform for Student Startups","prob":"University students often struggle to obtain initial funding for their startup ideas due to the absence of accessible and trustworthy financial platforms. Traditional investment systems do not cater to student entrepreneurs and lack transparency, trust, and data-driven decision support. On the other hand, potential lenders hesitate to invest because they cannot assess the legitimacy, feasibility, or success potential of a proposed idea. As a result, many innovative student startups fail to take off despite their potential.\r\nEdu-Vest aims to solve this problem by creating a secure, transparent, and intelligent platform that connects students with verified lenders. The system ensures mutual trust through blockchain-based record keeping and AI-driven evaluation modules that validate project cost, estimate equity, predict success, and generate immutable digital funding agreements stored securely on the blockchain. This enables both parties to make informed and confident investment decisions.","description":"Edu-Vest is a mobile application designed to connect verified students seeking startup funding with potential lenders through a secure, transparent, and data-driven platform. The system provides an organized structure where students can create startup proposals, specify budgets, and request financial support.\r\n\r\nOnce a project is submitted, the system validates whether the demanded cost is reasonable and uses predictive algorithms to estimate the likelihood of project success. Lenders can explore verified projects, view system-generated analytics, and decide whether to invest. When a lender agrees to fund a project, the system automatically generates a digital funding agreement containing details such as project name, funding amount, estimated equity, repayment or ownership terms, and duration. This agreement is securely stored on the blockchain as an immutable record, ensuring full transparency and protection against tampering.\r\n\r\nAn admin module handles verification, platform moderation, and analytics. A trust score mechanism evaluates user credibility based on activity and performance. The chat module allows students and lenders to communicate directly and finalize terms efficiently.\r\n\r\nBy integrating blockchain technology, machine learning models, and secure communication features, Edu-Vest aims to provide a transparent, intelligent, and reliable ecosystem that promotes student entrepreneurship and reduces investment risk.$$\nAdmin Module\r\nManages the overall system by verifying users, approving startup proposals, moderating content, and generating system reports. It ensures data authenticity and provides the administrator with analytics about users, projects, and investments.\r\n\r\nStudent Module\r\nEnables students to register, create and manage startup proposals, view analytics like cost validation and success prediction, and monitor investment offers. Once funded, students can access the automatically generated funding agreement that is stored on the blockchain.\r\n\r\nLender Module\r\nAllows lenders to browse verified startups, view detailed project data, assess estimated cost, success prediction, and equity percentage. Upon funding, a digital agreement is generated, providing full details of the funding plan, which is securely stored on the blockchain.\r\n\r\nBlockchain Module\r\nStores cryptographic hashes of all critical transactions and automatically generated funding agreements. This ensures immutability and trust by preventing unauthorized modification of records or agreements.\r\n\r\nAutomated Project Suggestor\r\nSuggests relevant startup proposals to lenders based on their interests, funding history, and project characteristics, helping them identify suitable investment opportunities.\r\n\r\nEstimated Cost Validation Module\r\nEvaluates whether the student\u2019s requested funding amount is reasonable and aligns with the project\u2019s description and category. This ensures fairness and prevents inflated demands.\r\n\r\nSuccess Prediction Module\r\nPredicts the likelihood of a project\u2019s success by analyzing data such as funding requirements, trust scores, project description, and other attributes. It assists both students and lenders in risk assessment.\r\n\r\nEquity Estimation Module\r\nCalculates the equity share a student should offer based on the funding amount, risk factor, and predicted success rate. It ensures balanced benefit for both parties.\r\n\r\nTrust Score Module\r\nAssigns a credibility score to users based on communication activity, successful project completions, and reliability. It helps create a transparent and trusted community.\r\n\r\nChat Module\r\nProvides a secure communication channel between students and lenders. It supports real-time messaging and helps finalize agreement details before the funding is locked.$$\nAdmin Module\r\n\r\nBlockchain Module\r\n\r\nSuccess Prediction Module\r\n\r\nEquity Estimation Module\r\n\r\nTrust Score Module\r\n\r\nResponsibilities:\r\n\r\nDevelop the admin dashboard for system control, user verification, and data management.\r\n\r\nImplement blockchain functionality for storing funding agreements and transaction hashes.\r\n\r\nBuild machine learning models for project success prediction and equity estimation.\r\n\r\nDesign algorithms to calculate dynamic trust scores based on user behavior.\r\n\r\nEnsure secure and efficient integration between blockchain, predictive modules, and Firebase backend.$$\nStudent Module\r\n\r\nLender Module\r\n\r\nAutomated Project Suggestor\r\n\r\nEstimated Cost Validation Module\r\n\r\nChat Module\r\n\r\nResponsibilities:\r\n\r\nDesign and implement the user interface for both student and lender roles.\r\n\r\nDevelop modules for project creation, cost validation, and automated project recommendation.\r\n\r\nManage user authentication, data synchronization, and app state management.\r\n\r\nImplement secure real-time chat functionality and link it with project interactions.\r\n\r\nHandle integration of student\u2013lender agreements with blockchain storage and ensure smooth data exchange between frontend and backend.$$\n$$\nNULL$$\nAI-Based Project Evaluation (Cost Validation & Success Prediction)$$\nDynamic Equity and Trust Scoring System$$\nAoutomated system for suggesting projects to lenders of their interests","comments":"","isDraft":0,"status":2,"created_at":"2025-10-08 21:52:35","updated_at":"2025-10-20 09:55:43","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1017,"project_id":1416,"title":"Pockito - An Integrated Super-App for Communication, Payments, and Everyday Services","prob":"In the modern digital landscape, users are forced to manage multiple disconnected applications for daily activities such as messaging, mobile payments, shopping, and service bookings. This fragmentation results in inefficiency, reduced productivity, inconsistent user experiences, and increased security vulnerabilities. Users in developing regions are further disadvantaged by limited device storage, unstable internet connectivity, and exposure to insecure third-party apps.\r\nPockito addresses this real-world problem by unifying essential digital functions secure chat, payments, and service access into a single, privacy-focused ecosystem. It eliminates app-switching friction and data redundancy through an integrated design supported by two-factor authentication, end-to-end encryption, and token-based access control. Additionally, offline caching ensures reliability in low-connectivity areas, while real-time analytics and notifications empower both users and merchants with actionable insights.","description":"Pockito is a unified super-app designed to integrate communication, digital payments, and essential daily services into one secure and intelligent mobile platform. The system aims to eliminate the inefficiency of using multiple separate applications for chatting, transactions, and service access by providing users with a seamless, connected ecosystem.\r\nThe application allows users to register, log in, and communicate in real time using a built-in chat system that supports text, media sharing, and offline message synchronization. Through the integrated digital wallet, users can perform secure money transfers, make in-app payments, and scan QR codes for quick transactions using a sandbox payment gateway. Security is ensured through two-factor authentication (2FA), data encryption, and JWT\/OAuth2-based access tokens, protecting users\u2019 identities and financial information.\r\nA distinctive component of Pockito is its mini-app framework, which enables third-party services such as e-commerce, booking, and food delivery to operate inside the main app through secure webviews. This extensible model encourages ecosystem growth and offers users access to multiple services without installing extra apps.\r\nTo enhance usability and data resilience, SQLite caching is employed to store offline chat history and wallet data, ensuring continuity during internet disruptions. Additionally, Firebase Cloud Messaging (FCM) delivers real-time notifications for messages, transactions, and promotional offers. An admin analytics dashboard provides insights into user engagement, chat frequency, transaction records, and system performance metrics, supporting better decision-making and scalability evaluation.$$\nThe proposed system Pockito is designed as a modular, scalable, and secure super-app comprising multiple interconnected components.\r\n1. Authentication & Authorization Module\r\nThis module manages user registration, login, and secure access control. It includes default functionalities such as signup, login, logout, and password management, and extends them with advanced security mechanisms.\r\nKey features include Two-Factor Authentication (2FA) using OTP\/email verification, JWT\/OAuth2 token-based sessions, and role-based access control (RBAC) for users and administrators. Passwords and sensitive data are encrypted using modern hashing algorithms.\r\n2. Real-Time Chat & Communication Module\r\nThis module provides the core communication feature of Pockito. It supports one-to-one and group chat, text messaging, and media file exchange.\r\nMessages are encrypted end-to-end, ensuring privacy and data security. To maintain continuity, SQLite-based offline caching stores unsent messages locally and syncs them with the server when connectivity resumes. Users also receive push notifications for new messages through Firebase Cloud Messaging (FCM).\r\n3. Digital Wallet & Payment Module\r\nThe wallet is central to Pockito\u2019s financial operations. It allows users to add funds, make payments, transfer money, and scan QR codes for transactions. All financial activities are simulated via sandbox payment APIs (Stripe) for demonstration purposes.\r\nEach transaction is securely logged and associated with unique user tokens for tracking and verification. Data encryption ensures secure handling of balances and payment details.\r\n4. Mini-App Framework Module\r\nThis module introduces extensibility within Pockito by hosting third-party or built-in services, such as shopping, booking, and delivery apps, as \u201cmini-apps\u201d embedded in the main application.\r\nDevelopers can integrate new mini-apps via secure WebView APIs that provide controlled access to Pockito\u2019s features like payment, authentication, and notifications.\r\n5. Offline Caching & Synchronization Module\r\nTo support users with unreliable internet, this module uses SQLite local databases to temporarily store user data such as chat history, wallet records, and mini-app transactions. When the device reconnects, the system automatically synchronizes the local data with the cloud backend.\r\nThis ensures seamless user experience, minimal data loss, and improved application reliability in low-connectivity environments.\r\n6. Analytics & Dashboard Module\r\nThis module provides real-time usage analytics and administrative insights. It visualizes metrics such as chat frequency, transaction records, mini-app usage, and payment volume.\r\nAdministrators can view statistics through interactive charts and reports.\r\n7. Notification & Event Module\r\nThis component handles push notifications for messages, payments, and promotional offers using Firebase Cloud Messaging (FCM). It ensures timely communication and keeps users engaged. Notifications are categorized as system alerts, transactional updates, or marketing messages.\r\n8. Administration & System Management Module\r\nThe admin panel allows system administrators to manage users, view logs, approve or disable mini-apps, and monitor performance and security alerts. It also supports analytics visualization and configuration of FCM notifications.$$\nIn Pockito, I shall develop\r\n(i) Real-Time Chat & Communication Module\r\n(ii) Digital Wallet & Payment Module\r\n(iii) Mini-App Framework Module\r\n(iv) Notification & Event Module$$\nIn Pockito, I shall develop\r\n(i) Authentication & Authorization Module\r\n(ii) Offline Caching & Synchronization Module\r\n(iii) Analytics & Dashboard Module\r\n(iv) Administration & System Management Module$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":0,"status":2,"created_at":"2025-10-08 22:07:58","updated_at":"2025-10-20 09:56:30","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":983,"project_id":1354,"title":"AI-Powered Car Diagnostic and Maintenance Tracking System with Mobile and Web Integration","prob":"1.High Repair Costs Due to Late Fault Detection\r\n2. Lack of Accessible Diagnostics for Ordinary Drivers\r\n3. No Centralized Vehicle Service History\r\n4. Reactive vs. Predictive Maintenance\r\n5. Limited Remote Monitoring\r\n6. Unsafe Driving & Roadside Breakdowns\r\n7. Digital Divide in Automotive Tech","description":"This Final Year Project proposes the development of an AI-powered OBD-II diagnostic and maintenance ecosystem that integrates a mobile application, a backend server, machine learning models, and a web dashboard to provide real-time vehicle health monitoring and predictive maintenance insights. The system leverages OBD-II data from the vehicle to identify faults, monitor live sensor readings, maintain service records, and generate intelligent predictions regarding future failures.\r\n\r\nThe mobile application (Flutter-based) will serve as the primary interface for drivers. It will display real-time parameters such as engine RPM, vehicle speed, coolant temperature, and fuel efficiency, while also showing active fault codes in both English and Urdu, with integrated voice assistance for improved accessibility. The app will also feature CRUD functionality for maintaining a digital service history, generate oil change reminders, and provide trip logs with GPS-based tracking. Users will have the option to export their data in PDF\/CSV formats and share it with mechanics.\r\n\r\nThe backend system will handle data ingestion, storage, and communication between modules. It will store logs in both relational and time-series databases, manage user authentication, and push notifications such as maintenance reminders. The backend will also integrate with machine learning models that analyze historical logs and real-time parameters to detect anomalies, identify patterns of recurring faults, and provide predictive insights into probable future issues.\r\n\r\nAdditionally, a web dashboard will be developed for mechanics and administrators, enabling them to remotely monitor vehicle health, analyze service histories, and view predictive analytics. This ensures that not only individual drivers but also fleet managers and service providers can benefit from the system.\r\n\r\nIn summary, this project aims to create a localized, intelligent, and user-friendly diagnostic ecosystem that bridges the gap between raw OBD-II data and meaningful, actionable insights for drivers and mechanics alike.$$\nBackend Module:\r\n(1) Authentication & User Management, (2) OBD-II Data Ingestion & Processing, (3) Database Management, (4) Service History CRUD Module, (5) Trip Logging & GPS Tracking, (6) Notification & Reminder System, (7) Reporting & Export Module, (8) Mechanic\/Administrator Dashboard API, 9() Security Module\r\n\r\nMachine Learning Modules\r\n1. Data Preprocessing & Feature Engineering, 2. Anomaly Detection Module, 3. Predictive Maintenance Module, 4. Fault Cause Prediction Module, 3. Driver Behavior Analysis Module, 4. Model Training & Evaluation Module, 5. Model Deployment Module, 6. Analytics & Visualization Module$$\nBoth student will be working on the backend together as we have to then integrate this backend with the machine learning module so the work would be done by both students$$\nBoth student will be working on the Machine learning together as we have to then integrate this backend with the machine learning module so the work would be done by both students$$\n$$\n$$\nAccident Detection and Alert System using\r\nAccelerometer$$\nMechanic Location & Marketplace Features$$\nChatbot Implementation","comments":"","isDraft":1,"status":1,"created_at":"2025-10-06 14:26:02","updated_at":"2025-10-17 11:40:02","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":999,"project_id":1389,"title":"AI-Powered OBD-II Diagnostic and Maintenance Tracking System with Web Integration","prob":"Instead of waiting for the car in the workshop, mechanics can see live fault codes & logs remotely. \r\nTransport companies (taxis, delivery vans, buses) can monitor all vehicles in one dashboard \u2192 detect issues before breakdowns.\r\nOwners lose track of manual service logs.\r\nWeb dashboard keeps complete digital history, accessible anytime.\r\nA mechanic can push updates (like \u201coil changed today\u201d) from the web \u2192 app user gets a notification.\r\nTrends from ML analysis help mechanics identify recurring issues (like failing O2 sensors or overheating engines).","description":"Part (2\/2)\r\nThe web application serves as the centralized monitoring and management hub for the Smart OBD-II Car Diagnostic System, complementing the mobile app. While the mobile app provides real-time vehicle interaction for individual drivers, the web dashboard is specifically designed for mechanics, administrators, and fleet managers who require a broader and more analytical overview of vehicle performance.\r\n\r\nThrough a secure login system, users can access a centralized database containing vehicle profiles, service histories, and user information. The application visualizes real-time OBD-II data, including engine RPM, coolant temperature, speed, and fault codes, using interactive dashboards and dynamic charts. This helps mechanics diagnose issues remotely without needing direct car connections.\r\n\r\nThe web module also includes a service history management system, enabling users to store, update, and retrieve maintenance records. All data are automatically synchronized with the mobile app, ensuring consistency between drivers and mechanics. Integration with the machine learning module allows the web app to display predictive maintenance insights, such as anomaly detection, component wear analysis, and probable causes of faults, based on historical sensor patterns.\r\n\r\nThe Trip History Visualization feature leverages Exploratory Data Analysis (EDA) and Seaborn to analyze and visualize data like speed, distance, and fuel usage, revealing driving patterns and performance insights. The Predictive Maintenance Dashboard, developed using Chart.js or Recharts, presents machine learning predictions and alerts through a visually rich interface, offering real-time health monitoring and fault trend visualization.\r\n\r\nAdditionally, the web app supports map-based tracking, allowing fleet operators to monitor vehicle routes and behavior, and includes report generation tools for exporting diagnostic and maintenance summaries in PDF or CSV formats. Overall, the system transforms complex OBD-II data into actionable insights, promoting proactive vehicle maintenance, reducing downtime, and improving decision-making for long-term automotive health.$$\n1. Authentication Module, 2.Vehicle & User Management, 3. Real-Time Monitoring, 4.Service History Module, 5.Trip History Visualization(EDA, seaborn), 6.Predictive Maintenance Dashboard (Chart.js\/Recharts-frontend & ML-training),7.Generate PDF\/CSV reports.$$\nIn this project i will develop the authentication module and user management module with real time data monitoring and service history module which will be integrated with the app being developed by another group assigned to Dr Jamal$$\nIn this project i will develop and design an interactive frontend for the whole web app and will also work on modules like trip history visualization , predictive maintenance module that will also be integrated by the backend being developed by the other group.$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-10-08 12:46:20","updated_at":"2025-10-17 11:40:20","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1033,"project_id":1356,"title":"Alkhidmat 360 An Intelligent Multi-Module System for Social Welfare and Qurbani Automation","prob":"Traditional Qurbani donations require time-consuming physical visits, while the portal lets donors select animals, choose shares, and pay online securely, saving time and effort.The platform replaces manual, paper-based welfare processes with an intelligent, unified system that automates donations, ration and food distribution, orphan sponsorship, loan approvals, and event management. By integrating AI-based verification and NADRA authentication, it ensures fairness, transparency, and efficiency across all services. Overall, it empowers organizations to deliver faster, verified, and data-driven social support to deserving communities.","description":"Alkhidmat 360 a comprehensive web platform designed to streamline all social welfare operations under the Alkhidmat Foundation. The system integrates multiple modules including Qurbani, Donations, Fitrana, Ration and Food Distribution, Orphan Sponsorship, Easy Loan, and Events Management into a single digital ecosystem. It enables donors to contribute online, applicants to request aid or loans securely, and administrators to verify beneficiaries through NADRA, manage logistics, and generate transparent, auditable reports. Overall, the platform aims to enhance efficiency, transparency, and accessibility while reducing manual workload and ensuring that every donation or service reaches the right people at the right time through intelligent automation and real-time monitoring.$$\nIn Alkhidmat Asaan ijtami Qurbani, the modules are (i) Users registration and authentication, (ii) Donor-Facing Module(landing catalog), (iii) Basket \/card Checkout, (iv) Payments & Reconciliation, (v) Skin Collection Request System, (vi) Submission and Assignment, (vii) Beneficiary Management, (viii) fitrana collection(ix) Educational Resources, (x) Feedback and Support(xi) orphan regestration (xii)Ramdan rashan drive module (xiii) loan request module (xiv)events module$$\n(v) Skin Collection Request System\r\nAllows donors to submit online requests for animal skin pickup after Qurbani, enabling volunteer to collect and track contributions efficiently.\r\n\r\n(vi) Submission and Assignment Module\r\nManages task assignments by automatically distributing incoming requests (donations, collections, or aid) among available staff or volunteers\r\n\r\n(viii) Fitrana Collection\r\nFacilitates online Fitrana payments, automates allocation to eligible recipients, and ensures timely distribution before Eid.\r\n\r\n(ix) Educational Resources\r\nProvides access to awareness materials, guidelines, and digital learning content about Alkhidmat\u2019s programs and community welfare activities.\r\n\r\n(x) Feedback and Support\r\nAllows donors and beneficiaries to share feedback or raise support queries, helping the foundation improve services and transparency.\r\n\r\n(xi) Orphan Registration\r\nDigitally registers orphan children, verifies guardianship, and connects them with sponsors for continuous support$$\n(i) Users registration and authentication, (ii) Donor-Facing Module(landing catalog), (iii) Basket \/card Checkout, (iv) Payments & Reconciliation, (xii)Ramdan rashan drive module (xiii) loan request module (xiv)events module\r\nUser Registration & Authentication \u2013 Secure login for donors, admins, and beneficiaries with verified access.\r\nDonor-Facing Module \u2013 Shows donation options like Qurbani, Fitrana, Loans, and Welfare Drives.\r\n Basket \/ Cart Checkout \u2013 Lets donors add multiple causes and donate in one transaction\r\nPayments & Reconciliation \u2013 Handles secure payments, receipts, and financial tracking.\r\nRamadan Ration Drive \u2013 Manages Ramadan donations and verified ration distribution.\r\nLoan Request Module \u2013 Accepts online loan applications with automated risk scoring.$$\n$$\n$$\nLoan Request Module$$\nRamadan Ration Drive$$\nOrphan Registration, Fitrana Collection","comments":" $$ Enough modules have been added for approval $$ Advised to partner with organization for real data for testing","isDraft":0,"status":2,"created_at":"2025-10-18 15:37:14","updated_at":"2025-10-20 15:43:08","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1016,"project_id":1400,"title":"FloodWatch 360","prob":"Floods cause widespread destruction in Pakistan every year, leading to loss of lives, infrastructure damage, crop destruction, and economic instability. Traditional flood monitoring relies on manual observation or delayed reporting, which makes early response difficult. This FYP aims to predict and monitor floods using satellite data and AI models so that authorities and communities can take timely action, minimize damage, and save lives.","description":"The system will collect and process multi-source satellite imagery, including MODIS near-real-time flood maps, Sentinel-1 and Sentinel-2 optical and radar imagery, as well as environmental datasets such as rainfall, soil moisture, NDWI (Normalized Difference Water Index), and NDVI (Normalized Difference Vegetation Index). These datasets will be preprocessed to remove noise, align timestamps, and clip to the geographic boundaries of Pakistan. Machine learning models, including LSTMs, and CNN-based approaches, will then be applied to this processed data to predict potential flood events. The results will be visualized through a user-friendly dashboard, displaying flood-prone areas on an interactive map, highlighting risk levels, and providing early warning alerts. By integrating remote sensing indices, environmental data, and AI-based modeling, the system aims to provide accurate, near real-time flood prediction and monitoring, supporting timely disaster preparedness, resource allocation, and data-driven decision-making.$$\nData Collection & Preprocessing\r\n\r\nAcquire MODIS, Sentinel, and NASA Earthdata rainfall\/soil moisture datasets.\r\n\r\nClean, filter, and preprocess satellite images and tabular data (remove noise, align timestamps, clip to Pakistan).\r\n\r\nFeature Engineering & Dataset Creation\r\n\r\nExtract key features such as water presence, rainfall intensity, NDWI (Normalized Difference Water Index), and soil moisture levels.\r\n\r\nBuild a labeled dataset for training and validation.\r\n\r\nAI Model Development\r\n\r\nImplement and train ML\/DL models (Random Forest, LSTM, or CNN-based models) to classify flood vs. non-flood conditions.\r\n\r\nTune and evaluate models for accuracy, precision, and recall.\r\n\r\nPrediction & Risk Mapping\r\n\r\nGenerate daily flood probability maps using model predictions.\r\n\r\nHighlight high-risk areas on geospatial layers for visualization.\r\n\r\nWeb Dashboard & Alerts\r\n\r\nDevelop an interactive web dashboard (using React) to visualize results on a map.\r\n\r\nProvide downloadable reports and optional email\/SMS alerts for early warning.$$\nIn this project, I will develop the Web Dashboard & Alerts module. My work will focus on building an interactive web application that visualizes AI-generated flood predictions through geospatial maps, risk levels, and summary reports. The dashboard will also include an optional alert system for notifying users via email or SMS.\r\n\r\nAdditionally, I will add a Future Trends Visualization feature that presents historical rainfall, soil moisture, and temperature data through interactive graphs. This will help users understand long-term environmental changes and support better planning. My goal is to make the system clear, responsive, and valuable for decision-makers.$$\nIn this project, I will be responsible for developing the AI Model Development & Prediction module. My work will involve collecting and preprocessing satellite imagery and hydrological data, perform feature engineering, and creating a structured dataset for training. I will design, train, and optimize machine learning or deep learning models that can accurately predict flood occurrences based on rainfall, soil moisture, and other key indicators. I will also generate flood probability maps and provide the processed data to be visualized on the dashboard. My goal is to ensure the prediction system is reliable, accurate, and capable of near real-time performance.$$\n$$\n$$\nThe dashboard automatically identifies and highlights the regions with the highest flood probability for a selected day or period, allowing users to quickly see which areas need immediate attention.$$\nUse simple colors (low, medium, high) to make the map easier to read.$$\nWhen a user clicks on a region, the dashboard will display interactive graphs showing flood probability, rainfall, soil moisture, and other environmental trends for that specific area.","comments":"","isDraft":0,"status":2,"created_at":"2025-10-08 22:07:55","updated_at":"2025-10-20 09:53:20","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1022,"project_id":1403,"title":"SmartEstate - House Price Prediction and Buyer Recommendation System Using Machine Learning","prob":"The real estate market often lacks transparency and data-driven insights, making it difficult for buyers to estimate fair house prices or find properties within their budget. This project aims to solve this issue by developing a machine learning-based system that predicts house prices based on various factors (location, area, rooms, etc.) and recommends suitable houses for buyers based on their financial details such as income, monthly expenses, and savings. The system provides data-driven and unbiased predictions, helping users make smarter purchasing decisions.","description":"The project focuses on building an intelligent web-based application that predicts house prices and recommends properties based on a buyer\u2019s financial profile. The system will first train machine learning models (Linear Regression, Random Forest, XGBoost) using real estate datasets. Users can view predicted prices by entering house features such as location, area, and number of bedrooms.\r\n\r\nA unique feature of the system is its Buyer Recommendation Module, where a user can input financial details (monthly income, expenses, savings, loan eligibility, etc.). The system then suggests house price ranges or suitable houses that match the buyer\u2019s affordability profile.\r\n\r\nThe application will include interactive data visualizations, model performance comparison, and an easy-to-use interface built using Python\u2019s Flask or Streamlit framework.$$\n1. Data Collection & Preprocessing Module \u2013 Collect datasets, clean missing values, handle outliers, encode categorical features, and normalize data.\r\n2. Exploratory Data Analysis (EDA) Module \u2013 Generate visual insights (heatmaps, correlation graphs) to analyze relationships between house features and prices.\r\n3. Model Training & Evaluation Module \u2013 Train multiple regression models (Linear Regression, Random Forest, XGBoost). Evaluate and compare their accuracy using MAE, MSE, and R\u00b2.\r\n4. House Price Prediction Module \u2013 Takes house details (location, area, rooms, etc.) and predicts the price using the trained model.\r\n5. Buyer Recommendation Module (New Feature) \u2013 Takes buyer\u2019s financial data (income, expenses, savings, loan eligibility) and suggests affordable house price ranges or available houses matching their profile.\r\n6. Web Interface Module \u2013 Build an interactive web application using Flask\/Streamlit where users can input details, view predictions, and see graphical analytics.$$\nMuhammad Usman: I will develop the Model Training & Evaluation Module and Buyer Recommendation Module, which include implementing machine learning models, affordability calculations, and designing logic for recommending houses based on financial details.$$\nAnas Arshad: I will develop the Data Preprocessing, Web Interface, and Prediction Modules, which include cleaning datasets, building the front-end interface, integrating the ML model, and displaying prediction results dynamically.$$\n$$\nOnline Real-Estate House Price prediction & Visualization$$\n1. Buyer Affordability Recommendation Module$$\n2. Model Performance Comparison Dashboard$$\n3. Interactive Web Interface with Real-Time Prediction","comments":"","isDraft":0,"status":2,"created_at":"2025-10-09 22:03:06","updated_at":"2025-10-20 09:53:42","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":973,"project_id":1392,"title":"Lawgorithm: An AI-Powered Case\u2013Lawyer Matching Platform","prob":"Clients face significant difficulty in identifying and connecting with the most suitable lawyers for their specific legal cases. Existing methods rely on personal contacts, word of mouth, or generic online directories, which are often inefficient, non-transparent, and unreliable. This leads to wasted time, uncertainty, and sometimes poor legal representation.","description":"The proposed Final Year Project, Lawgorithm: An AI-Powered Case\u2013Lawyer Matching Platform, aims to address the challenge faced by clients in finding the most suitable lawyers for their legal matters. In the current landscape, individuals often depend on informal referrals, personal networks, or generic online directories, which are inefficient, unreliable, and lack transparency. This project introduces a systematic, AI-driven solution to streamline the process of lawyer selection.\r\n\r\nThe platform enables clients to register, log in, and securely submit details of their legal cases through an intuitive web-based interface. Once a case description is provided, the system utilizes Natural Language Processing (NLP) techniques, specifically Sentence-BERT embeddings combined with FAISS-based similarity search, to semantically analyze the client\u2019s case and retrieve the most relevant lawyer profiles. To ensure transparency, the system generates a Fit Score (0\u2013100) that quantifies the relevance between a case and a lawyer, supported by explainable reasoning tags such as practice area match, court experience, or language compatibility.\r\n\r\nLawyers can create and manage their professional profiles, including details of expertise, years of experience, court specialization, and fees. An administrative module verifies lawyer credentials, ensuring authenticity and reliability of the platform. Standard functionalities such as login, logout, authentication, and authorization are integrated to protect sensitive client and lawyer data.\r\n\r\nThe system further ensures data security through encryption, controlled access, and compliance with privacy guidelines. Evaluation will be performed using information retrieval metrics such as Precision@K, NDCG, and MAP, alongside user feedback on clarity and trustworthiness of recommendations.\r\n\r\nIn summary, this FYP provides a secure, intelligent, and explainable recommendation system that bridges the gap between clients and verified legal professionals, reducing inefficiency and enhancing trust in the process of lawyer selection.$$\n1. Authentication & Accounts\r\n\r\nThis module provides role-based access for clients, lawyers, and administrators. It also manages profile completion to ensure that each user maintains accurate and complete information before accessing the system.\r\n\r\n2. Case Intake\r\n\r\nThis module allows clients to submit their legal cases by specifying key details such as category, court level, budget, urgency, language, and description. These inputs are stored for further processing and lawyer matching.\r\n\r\n3. Lawyer Profiles & Verification\r\n\r\nThis module manages lawyer accounts, ensuring verified enrollment details. It captures essential attributes such as practice area, supported languages, fees, and professional history, which are later used in the matching process.\r\n\r\n4. AI Matching Engine\r\n\r\nThis is the core module that performs lawyer recommendations. It uses Sentence-BERT embeddings along with FAISS similarity search to analyze client case details and lawyer profiles. It generates an explainable Fit Score for transparency.\r\n\r\n5. Search & Browse\r\n\r\nThis module enables users to search and explore lawyers based on filters. The filters include city, court, practice area, and keyword search, making it easier for clients to find relevant lawyers.\r\n\r\n6. Admin Console\r\n\r\nThis module is designed for administrative control. It allows verification of lawyers, management of categories and courts, and provides the functionality to import or export demo data for testing and system setup.$$\n1. Authentication & Accounts:\r\nImplement secure login, logout, signup, and role-based access (Client, Lawyer, Admin).\r\nHandle profile completion workflows for both clients and lawyers.\r\nEnsure data protection with authentication tokens and session management.\r\n\r\n2.Case Intake:\r\nDevelop forms for clients to submit new cases.\r\nCapture key fields such as category, court level, budget, urgency, language, and case description.\r\nStore cases in the database, linked to client profiles.\r\n\r\n3.Lawyer Profiles & Verification:\r\nEnable lawyers to create and update professional profiles.\r\nManage details like verified enrollment number, practice areas, languages, fees, and past case history.\r\nIntegrate with the admin module for profile verification.$$\n1.AI Matching Engine:\r\nImplement the core recommendation logic.\r\nUse Sentence-BERT embeddings with FAISS for similarity search between case descriptions and lawyer profiles.\r\nGenerate explainable Fit Scores with reasoning tags (e.g., court experience, language match).\r\n\r\n2.Search & Browse:\r\nProvide clients with advanced filtering and browsing options.\r\nSupport filters by city, court, practice area, and keyword search.\r\nOptimize queries for fast retrieval and user-friendly results.\r\n\r\n3.Admin Console:\r\nDevelop admin dashboards for lawyer verification and system oversight.\r\nManage system entities such as categories, courts, and practice areas.\r\nProvide options for importing\/exporting demo data for testing and evaluation.$$\n$$\nE-Lawyer$$\nAI-Powered Matching Engine: Uses Sentence-BERT and FAISS to recommend lawyers with an explainable Fit Score.$$\nLawyer Verification: Ensures only verified and trusted lawyer profiles are visible to clients$$\nAdvanced Search & Filters: Allows filtering by city, court, practice area, and keywords for refined results.","comments":"","isDraft":0,"status":2,"created_at":"2025-10-02 00:40:53","updated_at":"2025-10-20 09:52:29","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}{"id":1051,"project_id":1397,"title":"DailyVeg: Smart Agriculture and Logistics Solution","prob":"The goal of DailyVeg is to make it easy for farmers to manage their farms and sell their agricultural produce. It does this through incorporation of intelligent technologies, insights from artificial intelligence and provision of direct market within the application. The application increases the monitoring of agriculture, improves decision making in agriculture and increases the efficiency of agricultural products supply in order to increase farm profitability and ensure that customers receive fresh products without any problems.\r\nAmong the goals are:\r\n\r\n\u2022\tProvide weather forecasts and weather alerts.\r\n\u2022\tAllow farmers to assess the quality of soil.\r\n\u2022\tEnable farmers to track crop history.\r\n\u2022\tConnect farmers to wholesale buyers and shopkeepers.\r\n\u2022\tIntegrated chatbot for farmers assistance.","description":"DailyVeg is a smart agriculture and logistics platform designed to connect farmers, retailers, and consumers in a transparent and efficient way. The traditional vegetable supply chain suffers from issues like unfair pricing, high dependency on middlemen, wastage of crops, and lack of freshness. Farmers often earn less while consumers pay more, creating a gap that DailyVeg aims to solve using digital technology.\r\n\r\nThe system allows farmers to register, list their daily harvest, set prices, and receive real-time insights about demand and market trends. Consumers can browse available vegetables through a mobile app, compare prices, place orders, and track deliveries in real time. This direct farmer-to-consumer model ensures farmers gain higher profits while customers receive fresh vegetables at lower prices.\r\n\r\nDailyVeg also integrates a smart logistics module that assigns orders to the nearest delivery vehicle, uses route optimization algorithms, and ensures timely delivery with minimal costs. Cold storage and quality checks help maintain freshness during transportation. The admin panel manages registrations, monitors transactions, generates reports, and analyzes demand and supply patterns for better decision-making.\r\n\r\nThe project uses modern technologies such as React Native or Flutter for the frontend, Node.js or Django for the backend, MySQL or MongoDB for databases, and AI algorithms for demand forecasting. Secure payment gateways and real-time notifications enhance usability and reliability.\r\n\r\nIn conclusion, DailyVeg offers a sustainable solution to agricultural supply chain challenges. It benefits farmers by increasing profits, consumers by ensuring affordable fresh produce, and the overall system by reducing waste and optimizing logistics. This project demonstrates how technology can transform agriculture into a smarter, fairer, and more efficient industry$$\ni) User Registration & Authentication\r\nFE-1: Account Creation & Management\r\n\u2022\tEasy and simple registration procedure for truck drivers, farmers, and store owners.\r\n\u2022\tFor safe user management and sign-in, utilize Firebase Authentication.\r\nii) Profile Management & Cloud Storage\r\n\u2022\tGive users permission to update their personal and agricultural information.\r\n\u2022\tUse a cloud database to safely store data.\r\n\t\r\niii) Weather Forecasting & Alerts\r\nFE-1: Real-Time Weather Updates\r\n\u2022\tUse an API (such as OpenWeatherMap) to retrieve current weather information.\r\n\u2022\tUpdates are displayed on the user dashboard.\r\nFE-2: Weather Forecasting\r\n\u2022\tThe prediction for the next few days or weeks should be displayed.\r\nFE-3: Alerts & Notifications\r\n\u2022\tWhen severe weather conditions occur, send push alerts.\r\n\u2022\tSet off notifications according to the weather.\r\niv) AI-Powered Soil & Nutrient Analysis\r\nFE-1: AI-Based Soil Analysis\r\n\u2022\tUtilize AI algorithms to analyze soil conditions.\r\n\u2022\tGive advice on how to enhance the soil.\r\nFE-2: Nutrient Recommendations\r\n\u2022\tMake recommendations for the best fertilizers and nutrients for crops.\r\n\u2022\tMake suggestions based on the condition of the soil.\r\nv) Farm Management Dashboard\r\nFE-1: Farm Data Monitoring\r\n\u2022\tDisplay weather, soil, and crop history data visually.\r\n\u2022\tUse graphs and charts to illustrate trends.\r\nvi) Financial & Resource Management\r\nFE-1: Financial and Resource Management\r\n\u2022\tKeep track of your earnings, expenses, and resources.\r\n\u2022\tProvide an easily understood financial summary to help in planning.\r\nFE-2: Reports & Insights\r\n\u2022\tCreate automated farm performance reports.\r\n\u2022\tGive useful advice for improved management of farms.\r\nvii) B2B Marketplace for Farmers & Shopkeepers\r\nFE-1: Marketplace for Buying and Selling Produce\r\n\u2022\tFarmers list items and bargain with consumers over prices.\r\n\u2022\tMake it easier for buyers and sellers to communicate directly.\r\nFE-2: Seller and Buyer Profiles\r\n\u2022\tUse ratings to manage the profiles of buyers and sellers.\r\n\u2022\tFor transparency, maintain records of transaction history.\r\nviii)Transaction Management\r\n\u2022\tIntegration of safe payment gateways, such as Stripe.\r\n\u2022\tProvide notifications and order tracking.\r\nix) Delivery & Logistics System\r\nFE-1: Driver Assignment and Route Optimization\r\n\u2022\tDrivers are assigned to deliveries according to their location.\r\n\u2022\tFor route optimization, use the Google Maps API.\r\nFE-2: Real-Time Tracking\r\n\u2022\tTurn on real-time delivery tracking.\r\n\u2022\tInform users with any updates to the delivery status.\r\nx) AI Chatbot for Farmer Assistance\r\nFE-1: Chatbot for Instant Feedback\r\n\u2022\tIntegrate a chatbot to answer questions about farming.\r\n\u2022\tProvide immediate responses via a mobile interface.\r\nFE-2: Personalized Recommendations\r\n\u2022\tUsing user data, offer personalized farming advice.\r\n\u2022\tProvide customized farm management advice.\r\nxi)Automated Alerts & Recommendations\r\n\u2022\tNotify farmers when nutrient levels drop below optimal thresholds.\r\n\u2022\tProvide basic recommendations for improving soil health based on sensor data.$$\ni) User Registration & Authentication\r\niii) Weather Forecasting & Alerts\r\nv) Farm Management Dashboard\r\nx) AI Chatbot for Farmer Assistance\r\nxi)Automated Alerts & Recommendations\r\nix) Delivery & Logistics System$$\niv) AI-Powered Soil & Nutrient Analysis\r\nii) Profile Management & Cloud Storage\r\nvi) Financial & Resource Management\r\nvii) B2B Marketplace for Farmers & Shopkeepers\r\nviii)Transaction Management$$\n$$\n$$\n$$\n$$\n","comments":"","isDraft":1,"status":1,"created_at":"2025-11-23 18:18:04","updated_at":"2025-11-24 15:28:02","isReviewDraft":1,"isInternalDraft":0,"isExternalDraft":0,"reviewedDate":null,"markedDate":null}
No. Project Title Reg. No. Name Date & Time Evaluation Status
1. RecoMart CIIT/FA22-BCS-166/WAH
CIIT/FA22-BSE-084/WAH
CIIT/FA22-BSE-099/WAH
 SYED SAIF ULLAH BUKHARI
MUHAMMAD HOBAIB
MUHAMMAD ABDULLAH
 Done Accepted  
2. Dark RAT: “A Remote Access Trojan with Pivoting and Network Visualization” CIIT/FA22-BSE-042/WAH
CIIT/FA22-BSE-044/WAH
 AHMAR BILAL
HURAIRA FARAZ KHAN
 Done Accepted  
3. eDevice Registration & Monitoring Portal (eDRMP) CIIT/FA22-BSE-043/WAH
CIIT/FA22-BSE-117/WAH
 EMAN AKRAM
SYED ZAIN UL ABDIEN
 Done Accepted  
4. AI-Based Intelligent Learning Portal for Matric Students. CIIT/FA22-BCS-042/WAH
CIIT/FA22-BCS-057/WAH
 AYESHA SADAQAT
KASHAF MUDASSAR
 Done Accepted  
5. FareSmart CIIT/FA22-BSE-091/WAH
CIIT/FA22/BSE/079
 IQRA NOOR
Sibgha Noor
 Done Accepted  
6. EduSaaS 360 – Complete 360 SaaS Platform for Academic Institutions CIIT/FA22-BSE-107/WAH
CIIT/FA22-BSE-116/WAH
 BILAL AFZAL
MUZAMMIL IJAZ
 Done Accepted  
7. TourMate CIIT/FA22-BCS-053/WAH
CIIT/FA22-BCS-130/WAH
 KENZUL EMAN
AYESHA SIDDIQA
 Done Accepted  
8. Visio Spark Online CIIT/FA22-BSE-083/WAH
CIIT/FA22-BSE-097/WAH
 SYED HAIDER ALI SHAH
HAMMAD NASIR
 Done Accepted  
9. TruckLink: Verified Goods Movers Marketplace CIIT/FA22-BSE-033/WAH
CIIT/FA22-BSE-040/WAH
 OBAID ULLAH BUTT
AYESHA KHALID
 Done Accepted  
10. VisionMart CIIT/FA22-BCS-143/WAH
CIIT/FA22-BCS-148/WAH
 AHAD ALI
MUHAMMAD NOUMAN AFZAL
 Done Accepted  
11. AI-Powered Storytelling: Automated Story, Animation & Voice for Educational Moral Lessons CIIT/FA22-BCS-031/WAH
 MUHAMMAD HAMZA
 Done Accepted  
12. IntraView AI CIIT/FA22-BCS-015/WAH
CIIT/FA22-BCS-151/WAH
 USMAN AKRAM
HAFIZ MUHAMMAD ABDUL REHMAN
 Done Accepted  
13. Echo Lens (Industrial Project) CIIT/FA22-BCS-055/WAH
CIIT/FA22-BCS-114/WAH
 SYED JAWAD ALI
SAAD SHAHZAD
 Done Accepted  
14. Orbi (Orbiting Your Life) CIIT/FA22-BSE-052/WAH
CIIT/FA22-BSE-062/WAH
 LAIBA KHALID
ALISHA HASSAN
 Done Accepted  
15. 3D Virtual Try-On (VTO) CIIT/FA22-BCS-129/WAH
CIIT/FA22-BCS-154/WAH
 MUHAMMAD OMAMA KHALIL
JUNAID AHMAD
 Done Accepted  
16. Disruptive Duck AI – Conversational Intelligence for Project and Task Automation (Industrial Project) CIIT/FA22-BCS-068/WAH
CIIT/FA22-BCS-131/WAH
 EBTESAM NAEEM AWAN
SAHAAB HAMID QURESHI
 Done Accepted  
17. SaamayAI CIIT/FA22-BSE-046/WAH
CIIT/FA22-BSE-047/WAH
 SAJJAD ALI SHAH
ALI HASSAN SHAHID
 Done Accepted  
18. SmartStock: Conversational Inventory Intelligence System CIIT/FA22-BCS-120/WAH
CIIT/FA22-BCS-146/WAH
 FATIMA MEHBOOB
YUMNA BINT-E-YASIR
 Done Accepted  
19. ArtHive CIIT/FA22-BCS-090/WAH
CIIT/FA22-BCS-094/WAH
 MAHA RASHID
MUHAMMAD ABDULLAH
 Done Accepted  
20. InterStock-Stock Market Learning Platform CIIT/FA22-BCS-041/WAH
CIIT/FA22-BCS-159/WAH
 ISHMAL FATIMA
MUHAMMAD JAWAD UR RAHMAN
 Done Accepted  
21. Smart Chrome Extension for End-to-End Automated Call Management in Vicidial CIIT/FA22-BCS-102/WAH
CIIT/FA22-BCS-106/WAH
 ADEEL EJAZ
ALI RAZA
 Done Accepted  
22. EduVest: An Equity-Funding Platform for Student Startups CIIT/FA22-BCS-043/WAH
CIIT/FA22-BCS-074/WAH
 MUHAMMAD UMAR MUSTAFA
MUHAMMAD MAHHAD ZAHEER
 Done Accepted  
23. Pockito - An Integrated Super-App for Communication, Payments, and Everyday Services CIIT/FA22-BSE-004/WAH
CIIT/FA22-BSE-030/WAH
 MALIK HAMMAD ALI
AMMAR HUSSAIN
 Done Accepted  
24. Trendify AI – Find, Launch, and Sell Smarter CIIT/FA22-BCS-137/WAH
CIIT/FA22-BCS-145/WAH
 MUHAMMAD USMAN YOUNAS
MUHAMMAD RAMZAN
 Done Accepted  
25. AI-Powered Car Diagnostic and Maintenance Tracking System with Mobile and Web Integration CIIT/FA22-BSE-050/WAH
CIIT/FA22-BSE-104/WAH
 HASNAIN ALI KHAN
MEHREEN ABBASI
 Done Accepted  
26. AI-Powered OBD-II Diagnostic and Maintenance Tracking System with Web Integration CIIT/FA22-BCS-008/WAH
CIIT/FA22-BSE-051/WAH
 USWA AMJAD
SARA AKMAL
 Done Accepted  
27. Alkhidmat 360 An Intelligent Multi-Module System for Social Welfare and Qurbani Automation CIIT/FA22-BSE-001/WAH
CIIT/FA22-BSE-028/WAH
 RAYAN BHATTI
HASSAAN KHALID
 Done Accepted  
28. FloodWatch 360 CIIT/FA22-BCS-123/WAH
CIIT/FA22-BCS-134/WAH
 HASNAIN EJAZ
HAIDER ALI
 Done Accepted  
29. SmartEstate - House Price Prediction and Buyer Recommendation System Using Machine Learning CIIT/FA22-BCS-004/WAH
CIIT/FA22-BCS-024/WAH
 MUHAMMAD USMAN
Anas Arshad
 Done Accepted  
30. Lawgorithm: An AI-Powered Case–Lawyer Matching Platform CIIT/FA22-BCS-104/WAH
CIIT/FA22-BCS-162/WAH
 ABDUL MOEEZ
MUHAMMAD ANAS AMJAD
 Done Accepted  
31. Pawsitive Connections – An AI driven Animal Welfare & Adoption App CIIT/FA22-BCS-128/WAH
CIIT/FA22-BSE-101/WAH
 SAMIA AIN
HAFSA GUL
 Done Accepted  
32. Workflow 360: AI-Powered Unified Project Management Platform with Intelligent Task Automation for Small Teams CIIT/FA22-BSE-071/WAH
CIIT/FA22-BSE-081/WAH
 MUHAMMAD AQIB ABDULLAH MUGHAL
MUHAMMAD AHMAD
 Done Accepted  
33. CityFlow360 – A Real-Time Platform for Reporting, Tracking, and Resolving Civic Issues Efficiently CIIT/FA22-BSE-008/WAH
CIIT/FA22-BSE-019/WAH
 MARYAM SAJID
ARHAMA TABEER
 Done Accepted  
34. AI-Powered Smart Camera: An Intelligent Mobile Application for Real-Time Photography Assistance & AI Editing. CIIT/FA22-BCS-076/WAH
CIIT/FA22-BCS-095/WAH
 HASNAIN BAKHT
ASHIR ABBAS
 Done Accepted  
35. Skinsaga – Your AI-Powered Skincare & Wellness Companion (Industrial Project) CIIT/FA22-BCS-103/WAH
CIIT/FA22-BCS-141/WAH
 ANAYA AZHAR
BAKHTAWAR MURTAZA
 Done Accepted  
36. Sakinah – The Inner Peace: An Islamic Wellness & Guidance App (Industrial Project) CIIT/FA22-BCS-069/WAH
CIIT/FA22-BCS-161/WAH
 SYED MUJTABA SHAH
KHUSH BAKHT NAWAZ
 Done Accepted  
37. TRAINET CIIT/FA22-BSE-086/WAH
CIIT/FA22-BSE-088/WAH
 MAHEEN RAZA
TEHREEM MIRZA
 Done Accepted  
38. DailyVeg: Smart Agriculture and Logistics Solution CIIT/FA22-BCS-034/WAH
CIIT/FA22-BCS-039/WAH
 HAROON IMRAN
BADAR SALEEM
 Done Accepted