ad1ttyya/Algerian-Forest-Fires-Predection

# 🔥 Algerian Forest Fire FWI Predictor This project is a **web-based machine learning application** that predicts the **Fire Weather Index (FWI)** using **Ridge Regression**. It is trained on the Algerian Forest Fires dataset and deployed using **Flask** as a lightweight backend. --- ## 📌 Project Summary Forest fires pose a significant environmental threat, and the **Fire Weather Index (FWI)** helps assess fire risk based on weather and environmental conditions. This project: - Applies Ridge Regression to predict the FWI - Uses environmental variables such as temperature, humidity, wind speed, etc. - Provides an interactive web interface for input and real-time prediction --- ## 🧠 Machine Learning Model - **Algorithm**: Ridge Regression (Linear Regression with L2 regularization) - **Features Used**: - Temperature - RH (Relative Humidity) - Wind Speed - Rain - FFMC, DMC, ISI (Fire danger indices) - Classes (Fire occurrence) - Region (encoded numerically) - **Preprocessing**: Standard scaling using `StandardScaler` - **Evaluation Metrics**: RMSE, MAE, R² (covered in notebooks) --- ## 🌐 Web App Functionality Built using **Flask**, the app provides: - An input form for users to enter weather and fire index parameters - Backend logic that loads the trained Ridge Regression model and scaler - Prediction output displayed on a rendered HTML template **Endpoints:** - `/` → Entry point to the form - `/predictdata` → POST route for form submission and prediction --- ## 📂 Repository Structure ``` ├── application.py # Flask application ├── templates/ │ ├── index.html # Input form │ └── home.html # Prediction result display ├── models/ │ ├── ridge.pkl # Trained Ridge Regression model │ └── scaler.pkl # StandardScaler used in preprocessing ├── requirements.txt # Project dependenc …