harshpratap2006/Algerian-Forest-fires

# 🔥 Algerian Forest Fires – FWI Prediction A complete, end-to-end Machine Learning project that predicts the **Fire Weather Index (FWI)** for Algerian forest regions. The project covers the full data-science lifecycle — from raw data exploration and cleaning, through feature engineering and model selection, to deployment as an interactive **Flask** web application — demonstrating how machine learning can support early wildfire risk assessment. --- ## 📋 Table of Contents - About the Project - Dataset - Project Structure - Machine Learning Pipeline - Tech Stack - Getting Started - Usage - Results --- ## 📖 About the Project ### Motivation Forest fires are among the most devastating natural disasters worldwide. In Algeria, where vast stretches of Mediterranean forest cover the northern regions, wildfires routinely destroy thousands of hectares each year, threatening biodiversity, local communities, and agricultural livelihoods. Early and accurate prediction of fire danger is therefore critical for disaster-preparedness planning, resource allocation for firefighting crews, and issuing timely public warnings. ### What Is the Fire Weather Index (FWI)? The **Fire Weather Index** is the final output of the Canadian Forest Fire Weather Index System — an internationally adopted standard for estimating wildfire danger. The FWI combines several sub-indices that account for fuel moisture and fire behaviour: - **FFMC** (Fine Fuel Moisture Code) — moisture content of surface litter; indicates ease of ignition. - **DMC** (Duff Moisture Code) — moisture content of loosely compacted organic matter; affects fuel consumption. - **DC** (Drought Code) — moisture content of deep compact organic layers; reflects long-term drought effects. - **ISI** (Initial Spread Index) — expected rate of fire spread; derived from wind speed and FFMC. - **BUI** (Buildup Index) — total available fuel for combustion; derived from DMC and DC. A higher FWI value signals greater fire intensity and, …