Stellar Verification Program - Exoplanet Prediction System.

A comprehensive end-to-end machine learning system for classifying exoplanets and predicting planetary radius using Kepler Objects of Interest (KOI) data.

🌟 Project Overview

The Stellar Verification Program uses advanced machine learning to:

1. Classify KOI signals - Distinguish confirmed exoplanets from false positives
2. Predict planetary radius - Estimate the planetary radius in Earth radii

📋 Project Structure

Stellar/
├── frontend/              # React.js web application
│   ├── public/           # Static assets
│   ├── src/              # React components
│   │   ├── components/   # Reusable components
│   │   ├── App.js        # Main app component
│   │   └── App.css       # Styling
│   └── package.json      # Frontend dependencies
├── backend/              # Flask REST API
│   ├── app.py           # Main Flask application
│   └── requirements.txt  # Python dependencies
├── ml_pipeline/          # ML models and pipeline
│   ├── preprocessing.py  # Data preprocessing
│   ├── inference.py      # Model inference
│   ├── training.py       # Model training
│   └── eda.py           # Exploratory Data Analysis
├── data/                 # Data directory (add your dataset here)
└── README.md             # This file

🚀 Getting Started

Prerequisites

• Python 3.8+
• Node.js 14+
• PostgreSQL or SQLite

Backend Setup

1. Navigate to backend directory:

cd backend

2. Create Python virtual environment:

python -m venv venv
source venv/Scripts/activate  # On Windows

3. Install dependencies:

pip install -r requirements.txt

4. Run Flask server:

python app.py

The API will be available at http://localhost:5000

Frontend Setup

1. Navigate to frontend directory:

cd frontend

2. Install dependencies:

npm install

3. Start development server:

npm start

The application will open at http://localhost:3000

📊 API Endpoints

Prediction

• POST /api/predict - Get classification and regression predictions

  {
    "koi_period": 10.5,
    "koi_impact": 0.5,
    "koi_duration": 5.2,
    "koi_depth": 500,
    "koi_steff": 5778,
    "koi_srad": 1.0,
    "koi_smass": 1.0
  }

History & Statistics

• GET /api/predictions - Get all predictions
• GET /api/predictions/<id> - Get specific prediction
• GET /api/statistics - Get aggregated statistics
• GET /api/health - Health check

🤖 ML Pipeline

Data Preprocessing

• Handles missing values through imputation
• Feature normalization using StandardScaler
• Feature engineering for composite metrics

Model Architecture

• Classification: Random Forest Classifier

  • Metric: F1-score, ROC-AUC

• Regression: Random Forest Regressor

  • Metric: RMSE, MAE

🎯 Features

Frontend

✅ Input validation with immediate feedback ✅ Interactive prediction interface ✅ Real-time visualization of results ✅ Prediction history tracking ✅ Statistical analytics dashboard ✅ Responsive design

Backend

✅ RESTful API with CORS support ✅ Input validation and preprocessing ✅ Real-time model inference ✅ Prediction history in database ✅ Statistical aggregation

📈 Performance Metrics

Classification Performance

• F1-Score: Measures precision and recall balance
• ROC-AUC: Area under the receiver operating characteristic curve

Regression Performance

• RMSE: Root Mean Squared Error
• MAE: Mean Absolute Error

🔧 Configuration

Create a .env file in the backend directory:

FLASK_ENV=development
FLASK_DEBUG=1
DATABASE_URL=sqlite:///exoplanet_predictions.db

📚 Data Requirements

The system expects the following features:

• koi_period - Orbital period (days)
• koi_impact - Impact parameter
• koi_duration - Transit duration (hours)
• koi_depth - Transit depth (ppm)
• koi_steff - Star effective temperature (K)
• koi_srad - Star radius (solar radii)
• koi_smass - Star mass (solar masses)

Target variables:

• koi_disposition - Classification target (CONFIRMED/FALSE POSITIVE)
• koi_prad - Planetary radius (Earth radii)

🧪 Testing

Run the frontend tests:

cd frontend
npm test

🐛 Troubleshooting

CORS Issues

If you encounter CORS errors, ensure the Flask backend is running and accessible at http://localhost:5000

Database Issues

Delete exoplanet_predictions.db and restart the Flask app to reset the database

Port Already in Use

• Flask: Change port in app.py to a different port
• React: Use PORT=3001 npm start

📝 Documentation

See ml_pipeline/README.md for detailed ML pipeline documentation.

👥 Team

TEAM : RUSTY. Avinash Jha Ayush Belwal Ayush Kirti Singh

📄 License

[Add license information]