Usman361-AI/Forest-Fire-Prediction-and-Analysis-using-Machine-Learning

Machine learning project using classification, regression, clustering, and SHAP explainability on the Algerian Forest Fires dataset. Models fire occurrence and Fire Weather Index (FWI) to support proactive wildfire management. # 🌲 Forest Fire Prediction and Analysis using Machine Learning This project applies classification, regression, clustering, and explainable AI techniques to the **Algerian Forest Fires Dataset** to understand and predict fire occurrences and fire weather conditions. --- ## 🔍 Project Highlights - ✅ **Binary Classification**: Predicts if a fire occurred (`fire` vs `not fire`) - 📈 **Regression Modeling**: Predicts **Fire Weather Index (FWI)** - 🧠 **Clustering**: Groups data using KMeans and Agglomerative Clustering - 📊 **Explainability**: Uses SHAP for feature importance visualization - 🔄 **Data Standardization**: Applies `StandardScaler` for uniform feature scaling --- ## 💻 Models Used ### 🧪 Classification - Logistic Regression - Random Forest Classifier - Support Vector Machine (SVC) ### 📊 Regression - Linear Regression - Random Forest Regressor - Gradient Boosting Regressor ### 🤖 Clustering - KMeans - Agglomerative Clustering --- ## 📦 Dependencies Install all requirements using: ```bash pip install -r requirements.txt ``` --- ## 📁 Dataset - **Source**: UCI Algerian Forest Fires Dataset Place the cleaned dataset as: ``` forest-fire-ml/ ├── Algerian_forest_fires_dataset_CLEANED.csv ``` --- ## 🚀 How to Run ```bash python forest_fire_analysis.py ``` --- ## 📈 Visualizations - PCA-reduced clustering visualizations - SHAP summary plots for feature importance - Bar plots for Random Forest feature importances --- ## 🧠 Author - **USMAN** – AI & ML Enthusiast | GitHub: [Usman361-AI] --- ## 📜 License This project is licensed under the MIT License - see the LICENSE file for details.