anulsasidharan/Algerian_Forest_fire_prediction

# 🔥 Algerian Forest Fire Prediction System This project is a machine learning-based web application that predicts the occurrence of forest fires in the Algerian region using meteorological data. It demonstrates a complete end-to-end ML pipeline, from data preprocessing and model training to deployment using Flask and a clean, intuitive web interface. ## 🌲 Objective The primary goal is to predict forest fire occurrences based on environmental features such as temperature, humidity, wind speed, and rainfall. The model can assist environmental authorities and disaster management teams in early fire detection and prevention planning. ### 🧱 Project Architecture Diagram The project follows a modular structure, integrating data processing, machine learning, and web deployment components ### 🔄 Workflow The workflow encompasses data handling, model training, and deployment phases:​ 1. Data Collection: - Obtain Algerian Forest Fires dataset. 2. Data Preprocessing: - Clean and preprocess data. - Feature selection and scaling using StandardScaler. 3. Model Training: - Train Ridge Regression model on preprocessed data. - Evaluate model performance (e.g., R² score). 4. Model Serialization: - Save trained model as 'model.pkl'. - Save scaler object as 'scaler.pkl'. 5. Web Application Development: - Develop Flask application (application.py). - Create HTML templates for user input and result display. 6. Deployment: - Deploy Flask app to a web server (e.g., AWS Elastic Beanstalk ). - Users input data via web interface to get predictions. ### AWS Deployment Archetecture ### 🧠 Key Features - **Data Preprocessing**: Handling missing values, feature selection, encoding, and normalization. - **Model Training**: Multiple machine learning models were evaluated (e.g., Logistic Regression, Random Forest, SVM), with the best one chosen based on accuracy and precision. - **Flask-Based Web App**: Users can input weather conditions and receive an instant prediction on whe …