Key points are not available for this paper at this time.
This is a study on natural disasters, with an emphasis on urban floods, which have occurred more often involving relevant socio-economic impacts. Among the measures aimed at minimizing the consequences of floods, the most noteworthy are the empirical models linked with new technologies in a promising way and with lower financial cost compared to the models. Thus, a flood forecast model with applicability improved by new technologies becomes a protagonist in reducing the impacts caused by floods. The aim of this study is to compare urban flood prediction models using data mining, geo-processing (GIS) and machine learning techniques. The results illustrate the feasibility of applying in urban areas from the case study, with the aim of contributing to the management of urban environment disasters. Predicting the occurrence of floods is fundamental for alerting the riverside population and for mobilizing the civil defense action teams. To do so, models for forecasting flooding were drawn up one day in advance, where the results obtained from the models built with the techniques for classifying the rules showed satisfactory results. Therefore, forecasting the occurrence of urban floods can be used as a tool to support decision-making to guide the actions of the public authorities in the management of this type of disaster. The knowledge presented seeks to contribute in the debates on socio-environmental issues and public policies. In addition, the results of this study can be replicated elsewhere and for other types of variable data.
Rodrigues et al. (Wed,) studied this question.