Sanogo15/Water-Pollution-XGBoost-Ghana

Explainable XGBoost model for water pollution prediction in Ghana using SHAP analysis. # Explainable XGBoost Model for Water Pollution Prediction in Ghana ## Objective This project develops an explainable machine learning model using XGBoost to predict water pollution levels and analyze the main environmental factors influencing predictions. ## Methodology The workflow includes: - Data preprocessing - XGBoost model training - Model evaluation - SHAP-based explainability analysis ## Technologies - Python - XGBoost - SHAP - Scikit-learn - Pandas - NumPy ## Project Structure - data/ : dataset files - notebooks/ : experiments and analysis - src/ : source code - results/ : model outputs and visualizations ## Installation Install the required libraries: pip install -r requirements.txt ## Reproducibility The notebook and scripts provided allow reproduction of the model training and explainability analysis.