🚀 Life Insurance Claim Prediction System

An end-to-end Machine Learning project that predicts insurance claim approval by combining fraud detection, risk segmentation, premium prediction, and claim approval models.

🔥 Features

• 🛑 Fraud Detection (Rule-based + ML-based)
• 📊 Risk Segmentation using Clustering
• 💰 Premium Prediction
• ✅ Claim Approval Prediction
• 🌐 Full-stack integration (Frontend + Flask API)

🧠 Tech Stack

• Frontend: HTML, CSS, JavaScript
• Backend: Python, Flask
• Machine Learning: Scikit-learn
• Deployment: Render

📁 Project Structure

Insurance_Claim_Prediction/
│
├── Backend/
│   ├── src/
│   │   ├── fraud.py
│   │   ├── risk.py
│   │   ├── premium.py
│   │   ├── approval.py
│   │   ├── pipeline.py
│   │   └── __init__.py
│   │
│   ├── models/
│   │   ├── fraud_model.pkl
│   │   ├── risk_model.pkl
│   │   ├── premium_model.pkl
│   │   ├── approval_model.pkl
│   │   └── scaler.pkl
│   │
│   ├── app.py
│   └── requirements.txt
│
├── Frontend/
│   ├── index.html
│   ├── style.css
│   └── script.js
│
└── README.md

⚙️ How It Works

1. User enters customer and transaction details

2. Data is sent to Flask API

3. Pipeline executes:

   • Fraud Check
   • Risk Prediction
   • Premium Calculation
   • Claim Approval

4. Result is returned and displayed on UI

🚀 Run Locally

1. Clone repo git clone https://github.com/ayush-gangwar-09/Insurance-Claim-Prediction-.git

2. Go to project folder cd insurance-claim-prediction

3. Install dependencies cd Backend pip install -r requirements.txt

4. Run backend python app.py

5. Open frontend Open index.html in browser

🌐 Deployment

• Backend deployed on Render
• Frontend can be deployed on Netlify

💡 Key Highlights

• Modular ML architecture (separate models + pipeline)
• Real-world insurance workflow simulation
• Clean UI with modern design
• Scalable and production-ready structure

📌 Future Improvements

• Add charts and visualization
• Add authentication system
• Convert frontend to React
• Deploy using Docker and AWS

👨‍💻 Author

Ayush Kumar

⭐ If you like this project, give it a star!