Arshnoor-Singh-Sohi/Forest-Fire-Prediction

A machine learning-based web application that predicts forest fire occurrences and Fire Weather Index values using meteorological data from Algeria. The system implements classification and regression models with Flask, scikit-learn, and MongoDB to provide real-time fire risk assessment for environmental monitoring and disaster prevention. # 🔥 Forest Fire Prediction System ## 📋 Table of Contents - Project Overview - Live Demo - Features - Dataset Description - Technical Architecture - Machine Learning Approach - Model Performance - Installation & Setup - Usage Guide - API Documentation - Project Structure - Future Improvements - Contributions - License --- ## 🌟 Project Overview The Forest Fire Prediction System is a comprehensive machine learning application designed to predict the likelihood of forest fires in specific regions based on meteorological data. Forest fires pose significant environmental, economic, and social threats, particularly in vulnerable regions like Algeria where this dataset originates. Early prediction and detection are crucial for minimizing damage and protecting both ecosystems and human lives. This project leverages the Algerian Forest Fires dataset, which contains data from two regions of Algeria (Bejaia and Sidi Bel-Abbes) collected during the period from June 2012 to September 2012. Using various meteorological measurements and derived fire weather indices, we've built machine learning models that can: 1. **Classification Task**: Predict whether a forest fire will occur (binary classification: fire/no fire) 2. **Regression Task**: Predict the Fire Weather Index (FWI), a numerical indicator of fire danger The system is deployed as a web application that provides an intuitive interface for users to input weather conditions and receive predictions on forest fire risk, enabling forestry departments, environmental agencies, and emergency services to take preventive actions. --- ## 🔴 Live Demo Experience the application live at: forest-fire-prediction-1aa0… --- ## ✨ Features - **Real-time Prediction**: Input current weather parameters to get immediate fire risk assessment - **Dual Prediction Models**: - Binary classification (fire/no fire) for direct risk assessment - FWI regression for detailed fire danger rating - **Interactive Web …