Efficient detection and prediction of flood severity using machine learning algorithm

Abstract Flowing water flooding is a characteristic wonder which devastatingly affects human existence also, financial misfortunes. There have been different methodologies in contemplating stream flooding through the improvement of control measures and counteraction for this characteristic wonder. The group model is involved in the expectation of water level in relationship with flood seriousness. Flood prediction uses the most recent advancements in the Internet of Things and AI for the robotized examination of flood information is helpful to forestall catastrophic events. Exploration results show that group learning gives a more solid device to anticipate flood seriousness levels. The exploratory outcomes demonstrate that the outfit getting the hang of utilizing the LSTM model and arbitrary woodland beat singular models with an affectability, particularity, and exactness of 71.4%, 85.9%, and 81.13%. Results are produced by group models and provide warning about future flooding.