Monitoring and prediction system for Algerian forest fires using satellite data and machine learning.
# Algerian Forest Fire Prediction
**Predicting the Fire Weather Index (FWI) using meteorological data with a Flask-based web interface.**
---
## Table of Contents
- Project Overview
- Features
- Tech Stack
- Directory Structure
- Installation
- Usage
- Workflow
- Model Training and Evaluation
- Dataset
- Future Improvements
- Contributing
- License
- Credits
---
## Project Overview
This project demonstrates an end-to-end machine learning solution for predicting the Fire Weather Index (FWI) based on the Algerian Forest Fires dataset from 2012. The primary goal is to provide a reliable prediction of fire risk, which is then served through an intuitive web UI built with Flask. The FWI is a key indicator of fire danger, and this tool helps in assessing the risk level by mapping the predicted FWI to a clear risk band.
### Demo
---
## Features
- **FWI Prediction**: Utilizes a Ridge Regression model to predict the Fire Weather Index.
- **Modern UI**: A user-friendly interface with client-side validation for a seamless experience.
- **Risk Assessment**: Translates the FWI prediction into an easy-to-understand risk level (Low, Moderate, High, Extreme).
- **Interactive Notebooks**: Includes Jupyter notebooks for exploratory data analysis and model training.
---
## Tech Stack
- **Python**: Core programming language.
- **Flask**: Web framework for the user interface.
- **Scikit-learn**: For machine learning model development.
- **NumPy & Pandas**: For data manipulation and analysis.
---
## Directory Structure
Click to expand
```
Predict-Forest-Fire/
├── application.py
├── requirement.txt
├── README.md
├── Datasets/
│ ├── Algerian_forest_fires_dataset.csv
│ └── algerian_forst_fires_updated_datatset.csv
├── Models/
│ ├── ridge.pkl
│ └── scaler.pkl
├── Notebooks/
│ ├── EDA_and_FeatueEngineering.ipynb
│ └── model_training.ipynb
├── templates/
│ ├── index.html
│ └── home.html
└── docs/
├── cost_function.png
├── Screenshot 2026-01-02 …