Abstract :

Forest fires are one of the most destructive natural disasters, causing severe damage to ecosystems, wildlife, and human life. With climate change increasing fire frequency and intensity, early prediction has become critically important. This project proposes a forest fire prediction system using data science and machine learning techniques. Historical fire data, meteorological parameters, and satellite-based observations are analyzed to identify fireprone conditions. Advanced machine learning models such as Random Forest and XGBoost are employed to predict the probability of fire occurrence. The system aims to provide accurate, timely predictions to assist forest departments in early warning and prevention strategies. Experimental results show improved prediction accuracy compared to traditional statistical approaches.