ayelefransi/Ethiopian-Road-Traffic-Accident-Severity-Prediction

# Ethiopian RTA Severity Predictor — Full Stack Web App A production-grade full stack application that predicts road traffic accident severity in Addis Ababa, Ethiopia using a Stacking Ensemble ML model. --- ## Tech Stack | Layer | Technology | |-----------|-----------------------------------| | Frontend | React 18, Recharts, Framer Motion | | Backend | FastAPI, Python 3.9+ | | ML Model | sklearn Stacking Ensemble | | Styling | Custom CSS with design system | --- ## Model Performance | Metric | Score | |-------------------|--------| | Test Accuracy | 90.82% | | ROC-AUC | 0.9822 | | F1 (Weighted) | 90.91% | | F1 (Macro) | 88.25% | | Precision (Weight)| 91.14% | | Recall (Weighted) | 90.82% | --- ## Quick Start ### Linux / macOS ```bash chmod +x start.sh ./start.sh ``` ### Windows **Using Batch:** ``` Double-click start.bat ``` **Using PowerShell:** ```powershell .\start.ps1 ``` ### Manual Start **Backend:** ```bash cd backend pip install -r requirements.txt uvicorn main:app --host 0.0.0.0 --port 8000 --reload ``` **Frontend:** ```bash cd frontend npm install npm start ``` --- ## Project Structure ``` rta_webapp/ ├── backend/ │ ├── main.py # FastAPI app │ ├── requirements.txt │ └── model/ │ ├── best_rta_model.pkl │ └── model_metadata.json ├── frontend/ │ ├── public/ │ │ └── index.html │ ├── src/ │ │ ├── App.js │ │ ├── index.js │ │ ├── index.css # Global design system │ │ ├── components/ │ │ │ ├── Navbar.js │ │ │ └── Navbar.css │ │ └── pages/ │ │ ├── Home.js # Landing page with metrics │ │ ├── Predict.js # Interactive prediction form │ │ ├── Dashboard.js # Charts and model analytics │ │ └── About.js # Technical documentation │ └── package.json ├── start.sh # Linux/macOS launcher ├── start.bat # Windows la …