Abdullahh69/Forest-Fire-Detection

This project contains the Exploratory Data Analysis (EDA) and implementation of Machine Learning algorithms on the Algerian forest fires dataset. The goal of this project is to predict future forest fires in Algeria by analyzing historical data and training machine-learning models. # 🌲 Forest Fire Prediction using Machine Learning This project is part of an **Applied Machine Learning group assignment** focused on predicting forest fires in Algeria using historical weather data and machine learning techniques. ## 👥 Team Members - Md Abdullah Mia (x7626122) - Md Ashikuzzaman Eshti (x5389574) - Abdullah Chaudhry (x7393734) - Muhammad Safi (x5032223) ## 📌 Objective To analyze and model the **Algerian Forest Fires dataset** using Exploratory Data Analysis (EDA) and a set of machine learning algorithms, with the goal of predicting forest fire occurrences. ## 📂 Dataset The dataset contains daily meteorological and fire data for two regions of Algeria — **Bejaia** and **Sidi Bel-Abbes**, from **June to September 2012**. ### Features include: - **Date**: Day/Month/Year - **Temperature**: Max temperature at noon (°C) - **Relative Humidity (RH)**: % - **Wind Speed (Ws)**: km/h - **Rain**: mm - **FWI System indices**: FFMC, DMC, DC, ISI, BUI, FWI - **Region** - **Class**: `fire` or `not fire` ## ⚙️ Project Structure ### 🔍 1. Exploratory Data Analysis (EDA) - Identifying missing data - Region-wise splitting - Class distribution - Monthly trend analysis - Correlation heatmaps - Temperature distribution ### 🧹 2. Data Preprocessing - Label cleaning - Missing value handling - Feature encoding - Scaling - Train-test split ### 📊 3. Machine Learning Models Implemented the following classifiers: - Logistic Regression - Decision Tree - Random Forest - XGBoost (XGBClassifier) ### ⭐ 4. Feature Importance Top important features for prediction: - FFMC - ISI - FWI - RH - DMC ## ✅ Results Each model was evaluated using **confusion matrices** and overall accuracy scores. The best performing model was chosen based on prediction accuracy and feature significance. ## 🧠 Conclusion Machine Learning provides a powerful tool for early forest fire prediction. This proactive approach aids forest management agencies in preventing environmental and economic damage. …