Ejup Rustemi
University of Tetova, North Macedonia
E-mail: ejuprustemi2002@gmail.com
ORCiD iD: https://orcid.org/0009-0003-5133-7834
Abstract
The annual occurrence of significant floods results in the destruction and disruption of lives and communities all over the world, as well as the loss of billions of dollars in economic value. The impact of these occurrences is only going to become more severe as the climate continues to change, populations continue to expand, and urbanization continues to increase; around the world, the risk of flooding is predicted to quadruple over the course of the next century.
Therefore, it should come as no surprise that efforts are currently being made to develop new technologies that will enable more precise flood forecasting and estimation, as well as assist communities in being better prepared and more resilient. Machine learning, which is a subset of artificial intelligence, is one of those technologies that has the potential to significantly alter the manner in which we model flooding. Artificial intelligence and machine learning play a significant role in this transformation.
Keywords: AI, flood, data, management, computer systems
Introduction
The deep learning models that are now available are neural networks (NN) that are quite complicated and involve a great deal of computing on multiple layers. A number of recent developments in artificial intelligence technologies have been made feasible by advancements in computational capacity, as well as improvements in the quality and quantity of data that can be used to “train” modeling systems.
Scientists have already made significant progress in the application of powerful artificial intelligence (AI) and machine learning (ML) models to decrease the impact of flood risk on communities. These models have been used for a variety of purposes, including evaluating flood defense requirements and forecasting events in areas with limited data.
The Google Flood Hub is one example; its machine learning-powered algorithms estimate floods up to seven days in advance by utilizing deep learning and public data sources. Following in its footsteps, Amazon has implemented machine learning projects in Spain. These projects include an early flood warning system and an endeavor to assist local farmers in increasing agricultural yields while simultaneously lowering their water footprint.
First, let’s take a look at what machine learning is and how it operates before we move on to discussing how it may help advance flood modeling.