SafeTrip: uma ferramenta para detecção e notificação de acidentes em rodovias brasileiras a partir do Twitter.

Abstract

The highway system is the main means of transportation in Brazil. Traffic accidents are quite common in this mode of transportation, and are one of the leading causes of death in the country. Social media profiles of the Federal High-way Police (PRF) and other sources of information help to alert drivers about road accidents that have occurred, in order to reduce the frequency of this incidence. However, accessing this information by reading it while driving is illegal and further increases the risk of accidents. Therefore, this work proposes to develop a study on publications on social networks, particularly Twitter, to create a classification model, based on machine learning, that is capable of classifying tweets about information about the occurrence or not of road accidents. The expected results include the tweet classification model, as well as a mobile application that can notify drivers about accidents reported on their route, in real time and through audio.