Advancing Flood Forecasting: An Ensemble Machine Learning Approach for Enhanced Prediction

Flood control using AI is one of the most crucial domains, and the implementation of practical measures to mitigate subsequent damage is important. In this research, a prediction model was developed to enhance flood management efficiency. Over the past few years, the occurrence of floods in Kerala, India, has resulted in substantial adverse impacts on the local population, infrastructure, and ecological systems. Data on precipitation rates was used. In this research we aim to apply the ensemble machine learning (ML) model to predict floods. Ensemble methods have achieved high performance in ML in recent years and provided better performance and solutions to complex and advanced problems. ML methods used are logistic regression (LR), support vector machine (SVM), K-nearest neighbor (KNN), Naive Bayes (NB), and AdaBoost. The model was chosen based on the three best algorithms?LR, SVM, and AdaBoost?to build the ensemble model, and the final prediction was made using the ensemble methods (soft voting) technique. The ensemble model proved to perform better than single models and previous research, with yields of 98% accuracy.