Final Year Project 2025
Member 1 Member 2 Member 3 Member 4
Name Muhammad Muhamamd
Zeeshan Sadiq Mouazam Malik
Reg. No. 22-ARID-949 22-ARID-3161
Contact No. 0319-7362455 03709630840
Session(M/E) M M
FYP Idea 1:
Title: “Augmented Reality-based E-commerce Platform”
Description:
The Augmented Reality-based E-commerce platform is a web application that provides an immersive
shopping experience for customers. The platform allows users to interact with products in 3D, try out
clothing and accessories virtually, and see how furniture and decor would look in their own space. The
platform uses machine learning to provide personalized product recommendations based on user
behavior and preferences.
Key Features
Virtual Try-On: Users can try out clothing, accessories, and beauty products virtually using their
smartphone cameras.
3D Product Models: Products are displayed in 3D models, allowing users to rotate, zoom, and
interact with them.
AR Room Visualization: Users can see how furniture, decor, and other products would look in
their own space using AR technology.
Product Comparison: Users can compare different products side-by-side, with AR features
highlighting key differences.
Personalized Recommendations: The platform uses machine learning to provide personalized
product recommendations based on user behaviour and preferences.
Social Sharing: Users can share their AR experiences on social media platforms.
Secure Payment Gateway: The platform integrates a secure payment gateway for seamless
transactions.
Tech Stack:
Frontend: HTML, CSS3, JavaScript, AR.js, A-Frame, Three.js
Backend: Python Django / Node.js (with Express.js)
Database: MongoDB/ PostgreSQL
Machine Learning: Tensor Flow, PyTorch
3D Modeling: Blender, Maya
Benefits
Enhanced Customer Experience: The AR-based e-commerce platform provides an immersive
and interactive shopping experience, increasing customer engagement and satisfaction.
Increased Conversions: By allowing users to try out products virtually, the platform can increase
conversions and reduce returns.
Competitive Advantage: The platform's AR features can differentiate it from competitors and
establish a market leader position.
Target Audience
Demographics: Online shoppers aged 18-45
Psychographics: Tech-savvy individuals who value immersive experiences and are interested in
trying out new products virtually.
FYP Idea 2:
Title: “AI-Powered Personal Finance Manager with Smart Spending Insights”
Description:
Many people use apps like Mint, YNAB, or expense trackers — but most don’t explain why users
overspend, or recommend actions based on patterns. They also lack intelligent insights like predicting
upcoming expenses or financial risks.
Your project solves these problems by offering smart, personalized insights into user behavior, predictive
analytics, and clear visualizations.
Key Features:
1. User Panel:
• Upload or enter transaction history (CSV or manual)
• Categorize and track expenses/income
• Smart Insights (e.g., “You spend 20% more on weekends”)
• Spending Forecast (predicts future spending for the month)
• Budget planning and overspending alerts
• Visual reports (monthly, category-wise)
2. Admin Panel:
• Manage category rules
• Analyze common user behaviors
• Moderate or adjust smart recommendations
Machine Learning Component:
• Time-Series Forecasting: Predict future spending using past data (ARIMA, LSTM, or scikit-
learn models)
• Anomaly Detection: Flag unusual spending
• Clustering: Group users into spending profiles and adapt suggestions
• NLP: Parse transaction descriptions to categorize them automatically
Drawbacks It Fixes in Current Solutions:
Existing Apps Lack Your App Adds
Predictive analytics ML-based forecasts
Personalized spending tips Behavioral insights (e.g., mood, day, habit)
Open-source or academic availability Built in Django, customizable
AI explainability Simple explanations for predictions (e.g., SHAP or rules)
Tech Stack:
• Backend: Django + Django REST
• Frontend: React.js or Django Templates
• ML: scikit-learn, pandas, statsmodels, Prophet or TensorFlow
• Database: PostgreSQL
• Visualization: Plotly/Dash/Chart.js
• Deployment: Render, Heroku, or Vercel
Stretch Goals:
• Bank API integration (Plaid mock or dummy data)
• Goal-based budgeting (e.g., “Save for vacation”)
• Notification system via email/SMS
Machine Learning Component:
• Time-Series Forecasting: Predict future spending using past data (ARIMA, LSTM, or scikit-
learn models)
• Anomaly Detection: Flag unusual spending
• Clustering: Group users into spending profiles and adapt suggestions
• NLP: Parse transaction descriptions to categorize them automatically
Drawbacks It Fixes in Current Solutions:
Existing Apps Lack Your App Adds
Predictive analytics ML-based forecasts
Personalized spending tips Behavioral insights (e.g., mood, day, habit)
Open-source or academic availability Built in Django, customizable
AI explainability Simple explanations for predictions (e.g., SHAP or rules)
Tech Stack:
• Backend: Django + Django REST
• Frontend: React.js or Django Templates
• ML: scikit-learn, pandas, statsmodels, Prophet or TensorFlow
• Database: PostgreSQL
• Visualization: Plotly/Dash/Chart.js
• Deployment: Render, Heroku, or Vercel
Stretch Goals:
• Bank API integration (Plaid mock or dummy data)
• Goal-based budgeting (e.g., “Save for vacation”)
• Notification system via email/SMS
Submission date:
Submitted To:
