KodewithArun/Algerian-Forest-Fire-Prediction-System

A machine learning web application that predicts the **Fire Weather Index (FWI)** — a critical measure of forest fire risk — based on weather and environmental conditions in two regions of Algeria: **Bejaia** and **Sidi Bel-Abbes**. # 🌲 Algerian Forest Fire Prediction System A machine learning web application that predicts the **Fire Weather Index (FWI)** — a critical measure of forest fire risk — based on weather and environmental conditions in two regions of Algeria: **Bejaia** and **Sidi Bel-Abbes**. This project combines **data cleaning**, **exploratory data analysis (EDA)**, **regression modeling**, and a **Flask-based web interface** to deliver real-time fire risk predictions. --- ## 📌 Table of Contents - Project Overview - Dataset - Key Features - Exploratory Data Analysis (EDA) - Model Training - Flask Web Application - How to Run the App - Folder Structure - Dependencies --- ## 🚀 Project Overview This system predicts the **Fire Weather Index (FWI)** using a **Ridge Regression model** trained on real weather data collected from June to September 2012. The FWI is a continuous index that reflects the potential for fire spread and intensity. While the original dataset includes a binary label (`fire` / `not fire`), this project treats **FWI prediction as a regression task**, enabling more granular fire risk assessment. The trained model is deployed via a **Flask web application**, where users can input weather parameters and get an instant FWI prediction. --- ## 📂 Dataset - **Name**: Algerian Forest Fires Dataset - **Source**: Kaggle / UCI ML Repository - **Time Period**: June – September 2012 - **Instances**: 244 (122 from Bejaia, 122 from Sidi Bel-Abbes) - **Features**: - `Temperature`: Temperature in °C - `RH`: Relative Humidity (%) - `Ws`: Wind Speed (km/h) - `Rain`: Rainfall (mm) - `FFMC`, `DMC`, `DC`, `ISI`, `BUI`, `FWI`: Components of the Canadian Forest Fire Weather Index (FWI) System - `Classes`: 0 = "not fire", 1 = "fire" - `Region`: 0 = Bejaia, 1 = Sidi Bel-Abbes - **Target Variable**: `FWI` (predicted as a continuous value) --- ## 🔍 Key Features - Data cleaning and region-based splitting - Removal of multicollinear features (correlation > 85%) - Feature scaling us …