SergeMoya/AI-malaria

A machine learning system achieving 82.1% accuracy in predicting malaria incidence across African regions. Features include Random Forest ensemble with temporal cross-validation, automated data preprocessing pipeline, and epidemiological pattern analysis. Processes health data from 50 countries to optimize disease prevention strategies. # Malaria Prevention Analysis in Africa This project analyzes the effectiveness of preventive measures against malaria in African countries using machine learning techniques. ## Project Structure ``` malaria/ ├── malaria_analysis.py # Main analysis script ├── requirements.txt # Python dependencies ├── DatasetAfricaMalaria.csv # Input dataset (not included) └── output/ # Generated visualizations ├── prevention_effectiveness_heatmap.png └── prediction_accuracy_plot.png ``` ## Setup and Installation 1. Create a virtual environment (recommended): ```bash python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate ``` 2. Install required packages: ```bash pip install -r requirements.txt ``` ## Usage 1. Place your `DatasetAfricaMalaria.csv` file in the project root directory 2. Run the analysis script: ```bash python malaria_analysis.py ``` 3. Check the `output` directory for generated visualizations: - `prevention_effectiveness_heatmap.png`: Shows correlation between preventive measures and malaria incidence - `prediction_accuracy_plot.png`: Displays model prediction accuracy with R² and RMSE metrics ## Dataset Format The input CSV file should contain the following columns: - country: Name of the African country - bed_nets: Data about bed net usage - antimalarial_medication: Data about antimalarial medication usage - malaria_incidence: Target variable showing malaria cases ## Output The script generates: 1. A heatmap showing the effectiveness of different prevention methods across countries 2. A prediction accuracy plot comparing actual vs. predicted malaria incidence 3. Printed performance metrics (R² and RMSE) ## Error Handling The script includes error handling for: - Missing input file - Data loading issues - Visualization creation errors - Model training problems