hathimds/Algerian-Forest-Fire-ML-Prediction

End-to-end ML project to predict forest fires in Algeria using weather indices. Includes EDA, feature engineering, Ridge Regression, and Flask deployment. # 🌲 Algerian Forest Fire – Fire Weather Index (FWI) Prediction This project presents a complete machine learning pipeline to predict the **Fire Weather Index (FWI)** using real-world meteorological data from Algerian forest fire records. The goal is to support early warning and forest fire risk assessment through accurate FWI predictions. --- ## 📌 Project Overview - Cleaned and processed meteorological data relevant to fire weather conditions. - Built a **Ridge Regression** model to predict FWI based on weather features. - Developed a **Flask web application** that allows users to input meteorological parameters and receive predicted FWI values in real time. - Successfully deployed the web app on **Render** for public access. --- ## ✅ Key Features - Data preprocessing including cleaning, handling missing values, and feature engineering. - Ridge Regression model trained and evaluated for robust prediction. - Interactive Flask interface for user-friendly input and output display. - Cloud deployment enabling easy access from anywhere. --- ## 🗂 Dataset The dataset includes meteorological variables such as temperature, humidity, wind speed, rainfall, and others recorded during Algerian forest fire events. --- ## 🔍 Data Processing & Modeling Steps - Data cleaning and preprocessing - Exploratory data analysis - Feature selection - Model building using Ridge Regression - Model evaluation with relevant metrics - Web app development for prediction interface - Deployment on Render platform --- ## 🛠 Technologies & Tools - **Python** - **Flask** - **Scikit-learn** - **Pandas, NumPy** - **Render (for deployment)** --- ## 🌐 Model Deployment - Deployed the Flask app on **Render** **Live App:** Link --- ## 📬 Connect with Me Feel free to reach out or explore more of my projects on LinkedIn: Haathim Manaf - LinkedIn --- ## 📜 License This project is licensed under the MIT License.