Real-Time Urban Flood Forecasting And Alert System

Flood prediction is critical for mitigating the devastating impacts of floods on communities and infrastructure. This project aims to develop a machine learning-based model to predict the probability of flooding in a given region using various environmental and socio-economic factors. The dataset used for this project includes 21 features such as monsoon intensity, deforestation, climate change, and urbanization, with the target variable being FloodProbability, a continuous value between 0 and 1. We apply four different regression algorithms—Linear Regression, XGBoost Regressor, Random Forest Regressor, and Stacking Regressor—to predict the flood probability. Data preprocessing techniques such as feature scaling and data splitting are used to prepare the data for model training. The models are evaluated using performance metrics like Mean Squared Error (MSE) and R² score. Among the models, the Stacking Regressor and Linear Regression show the best performance, achieving an R² of 0.8449. This project highlights the potential of machine learning models to improve flood prediction accuracy, providing a useful tool for disaster preparedness and resource allocation.