Dhairya-45/Algerian-Forest-Fire-Prediction

# 🔥 Algerian Forest Fire Prediction An end-to-end Machine Learning web application to predict the **Fire Weather Index (FWI)** using meteorological data from the Algerian Forest Fires dataset. 🌐 **Live Demo:** Click Here > ⚠️ Hosted on Render's free tier — first load may take 50+ seconds to wake up. --- ## 📌 Problem Statement Forest fires cause massive environmental and economic damage. This project predicts the **Fire Weather Index (FWI)** based on weather and fire-related features to estimate fire risk levels. --- ## 🚀 Features - Real-time Fire Weather Index (FWI) prediction - Fire risk classification (Low, Moderate, High, Extreme) - User-friendly Flask web interface - Scalable ML pipeline --- ## 🧠 Machine Learning Pipeline 1. Data preprocessing & cleaning 2. Feature selection 3. Feature scaling using **StandardScaler** 4. Model training using **Ridge Regression** 5. Model serialization using **Pickle** --- ## 🛠️ Tech Stack - **Programming Language:** Python - **Machine Learning:** Scikit-learn - **Web Framework:** Flask - **Frontend:** HTML, CSS - **Deployment:** Render - **Version Control:** Git & GitHub --- ## 📂 Project Structure ``` Algerian-Forest-Fire-Prediction/ │ ├── model/ # Jupyter notebooks for EDA & model training ├── pkl/ # Serialized model and scaler files │ ├── ridge.pkl │ └── scaler.pkl ├── templates/ # HTML templates │ └── home.html ├── application.py # Flask application ├── requirements.txt # Python dependencies └── README.md ``` --- ## 📊 Dataset - **Source:** UCI Machine Learning Repository - Algerian Forest Fires Dataset - **Regions:** Bejaia (northeast Algeria) and Sidi Bel-abbes (northwest Algeria) - **Period:** June to September 2012 ### Input Features | Feature | Description | |---|---| | Temperature | Temperature at noon (°C) | | RH | Relative Humidity (%) | | Ws | Wind Speed (km/h) | | Rain | Total rainfall (mm) | | FFMC | Fine Fuel Moisture Code | | DM …