Neilla513/Climate-and-Health-Resilience-In-Cameroon-IndabaX-Cameroon-2026

🌍 AI for Climate & Health Resilience | IndabaX Cameroon 2026 Predictive modeling of air quality health risks from meteorological data across 42 Cameroonian cities. Includes full pipeline, trained AI model, interactive multilingual dashboard, and geospatial insights. # Climate-and-Health-Resilience-In-Cameroon-IndabaX-Cameroon-2026 🌍 AI for Climate & Health Resilience | IndabaX Cameroon 2026 Predictive modeling of air quality health risks from meteorological data across 42 Cameroonian cities. Includes full pipeline, trained AI model, and geospatial insights. # 🌍 IndabaX Cameroon 2026 – Climate & Health Resilience **L'IA au service de la résilience climatique et sanitaire au Cameroun** ## 🎯 Project Overview This project develops an **AI-powered system** to predict **daily air quality health risk** across Cameroon using only meteorological data. It helps public health authorities anticipate respiratory risks caused by heat waves, wind stagnation, and low precipitation. **Key Innovation**: We created a **health-focused risk score** (0–100) that combines temperature, wind, precipitation, and solar radiation — directly linked to real-world respiratory health impacts. --- ## 📋 Deliverables (Hackathon Requirements) - ✅ **Prediction Engine** – Full documented pipeline + trained model - ✅ **Interactive Dashboard** – Multilingual Streamlit app with maps & scenario simulator - ✅ **Geospatial Analysis** – Risk maps for 42 cities - ✅ **Future Climate Simulation** – What-if scenarios (+2°C warming, etc.) - ✅ **Documentation** – Complete & reproducible --- ## 🛠️ Tech Stack - **Language**: Python 3.10+ - **Data Processing**: pandas, numpy, geopandas - **Modeling**: Random Forest + PyTorch (ready for upgrade) - **Visualization**: Plotly, seaborn, Folium - **Dashboard**: Streamlit (multilingual: EN / FR) - **Deployment**: Ready for Streamlit Cloud / Hugging Face Spaces --- ## 📦 Dataset - **Source**: Official IndabaX 2026 dataset (`Dataset_complet_Meteo.xlsx`) - **Size**: 87,240 daily observations - **Coverage**: 42 cities × 10 regions (2020–2025) - **Variables**: Temperature, wind, precipitation, evapotranspiration, solar radiation, weather code, etc. ---