Forest fires pose a significant threat to ecosystems, biodiversity, and human life, with increasing frequency and intensity due toclimate change. Early detection and accurate prediction are critical for effective response and mitigation. This paper presents a machinelearning-based approach to forest fire detection and risk prediction using environmental data such as temperature, humidity, wind speed,and rainfall. Various classification algorithms, including Random Forest, Support Vector Machine (SVM), and Logistic Regression, wereevaluated to classify fire-prone conditions. The model was trained and tested on publicly available datasets and achieved high accuracy andreliability in predicting potential fire outbreaks. Additionally, feature importance analysis highlighted the key factors influencing fire risk.The proposed system can be integrated with IoT sensors or satellite data feeds to enable real-time monitoring and early warning systems.The results demonstrate that machine learning techniques offer a robust, scalable, and cost-effective solution for forest fire managementand can significantly enhance proactive disaster response strategies.