An intelligence platform that predicts flood, storm, and heat risk using:
- Machine Learning (XGBoost)
- 7-day rainfall signal analysis
- Live global disaster alerts (ReliefWeb + GDACS)
- Real-time Open-Meteo weather API
| Feature | Technology |
|---|---|
| Flood ML Prediction | XGBoost |
| Live Weather Data | Open-Meteo |
| Real Disaster Alerts | ReliefWeb + GDACS |
| Hybrid Risk Engine | ML + Live Intelligence |
| Frontend | Streamlit |
| API Backend | FastAPI |
This system fuses AI + disaster intelligence to provide city-level climate risk classification:
Low — Medium — High.
The risk engine combines:
- Machine learning probability
- Rainfall accumulation
- National disaster alerts
- Local event detection (city)
- Temperature & wind hazard rating
Outputs:
- Flood risk
- Heat risk
- Storm risk
- AI Explanations
- Risk Escalation Logic
cd backend
pip install -r ../requirements.txt
uvicorn main:app --reload