The primary objectives of this project revolve around building an effective flood risk prediction system. Data collection and preprocessing focus on gathering meteorological and geographical data, followed by essential steps such as handling missing values, scaling features, and selecting the most relevant ones to ensure high-quality input for modeling. Model development involves implementing advanced machine learning techniques specifically TabNet, Stacking Classifier, and Voting Classifier to accurately predict flood risk levels from the prepared data. System development centers on creating a secure web-based platform that enables users to register, log in, input data, and obtain flood risk classifications using the trained models. Performance evaluation is conducted through standard metrics including accuracy, precision, recall, and F1-score to rigorously assess model reliability. The user interface is designed to be intuitive and user-friendly, utilizing HTML, CSS, and JavaScript to facilitate seamless interaction and clear interpretation of prediction results. Finally, model optimization is achieved through hyperparameter fine-tuning and the exploration of ensemble methods to further enhance overall prediction accuracy and robustness.