Analysis of sensitive urban form indicators of flood susceptible prediction based on machine learning models

BACKGROUND AND OBJECTIVES: Flooding is one of the biggest challenges affecting the economy and people's well-being. Various approaches have been employed in prior research to examine spatial data and enhance flood response effectiveness, such as the utilization of machine learning methodologies to aid in decision-making within the realm of urban planning. However, different machine learning models serve different purposes depending on their learning processes and computation techniques. The objective of this study is to create a machine learning algorithm that can evaluate areas prone to flooding. The study purpose is to offer valuable insights to city authorities and aid in urban planning, ultimately enhancing the safety and adaptability of the city's inhabitants METHODS:In order to evaluate flood risk areas and offer valuable insights for urban management and planning, a technique was devised. A total of eleven multiclass classification algorithms were employed to analyze eight urban factors, enabling the assessment of flood risk. The outcomes were then visualized on a map using geographic information systems. FINDINGS: The study discovered that the Bagging Decision Tree Algorithm model produced the best flood risk assessment model, with an accuracy of 88.58 percent compared to the government's flood simulation model results. Furthermore, precipitation, building coverage ratio, and floor area ratio emerged as the top three crucial factors influencing the likelihood of flooding in the city. CONCLUSION: The Bagging Decision Tree Algorithm model provides a thorough evaluation of flood risk, presenting crucial information for urban governance and development. Integrating key variables such as rainfall, building coverage ratio, and floor area ratio into flood risk management strategies is crucial for mitigating the impact of floods in economically significant urban areas.