Review of flood prediction hybrid machine learning models using datasets

Abstract

Floods are among the most destructive natural disasters, and they are extremely difficult to model. Over the last two decades, machine learning (ML) methods have made significant contributions to the advancement of prediction systems that provide better performance and cost-effective solutions by mimicking the complex mathematical expressions of physical flood processes. Because of the numerous benefits and potential of ML, its popularity has skyrocketed. Researchers hope to discover more accurate and efficient prediction models by introducing novel ML methods and hybridising existing ones. The main focus of this paper is to show the state of the art of hybridising ML models in flood prediction. The most effective strategies for improving ML methods are hybridization, data decomposition, algorithm ensemble, and model optimization.