Método de sensoriamento social para caracterização e detecção de eventos urbanos: uma aplicação em acidentes de trânsito

Abstract

Traffic accidents are a recurrent problem in urban areas, causing damages and injuries. Currently, there are several public and private entities that encourage the creation of solutions that help to minimize the occurrence of these accidents in urban areas. In this way, we present a solution, which uses the concepts of Social Sensing for monitoring and characterization of traffic accidents. Social sensing is a new paradigm in which is performed a distributed process of collecting social data, through people sharing contextual data voluntarily. As a case study, we applied the solution for monitoring and characterization of traffic in urban areas, as their inhabitants share a large number of information on social networks. Furthermore, there are official databases related to the city, made available by the government, which can be used for validation of the proposed solution. The solution considers the limitations of Social Sensing and the experiments use both official public data and social data from Twitter and Foursquare. The results show that, for the scenarios evaluated, it becomes possible to use social networks as an alternative of traffic accidents monitoring and characterizing.