Abstract

Abstract Floods continue to be one of the destructive natural disasters, disrupting millions of lives and causing most significant economic losses annually. The need for accurate and timely flood forecasting is critical for effective disaster prevention and mitigation. Traditional hydrological and statistical models struggle to represent the nonlinear relationships between rainfall, soil moisture, and river discharge. This paper proposes a machine learning (ML)-based flood forecasting framework that leverages historical and real hydrological data to enhance accuracy and adaptability. Supervised learning algorithms- Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression, and Multi-Layer Perceptron were implemented and evaluated. Among these, MLP demonstrated superior predictive performance due to its ability to capture complex nonlinear dependencies. The proposed model achieved higher accuracy and faster response time than traditional models, establishing a scalable and intelligent decision-support system for flood management