Mexico city traffic analysis based on social computing and machine learning

Nowadays artificial intelligence is immersed in all the people’s activities. Internet and mobile devices let us produce and consult information related to social and urban aspects. The crowd sourcing information and the social computing analyze and implement solutions to real world problems using the web content generated by social media and internet users. One of the urban factors that affect people’s activities is the vehicular traffic, every day traffic produces high stress levels and time delays when people are trying to move from one place to another using their cities highway. Vehicular traffic problems impact directly over the human’s health and over the financial dynamics of the affected cities. In the present approach, social computing is implemented by analyzing crowd sourcing information related to vehicular traffic, and computing regressions over the identified traffic events, to determine how traffic would affect an urban area at different hours. The consulted crowd sourcing information is obtained from Twitter. The traffic events forecast is implemented using a machine learning regression algorithm; the retrieved data from the social network and the regression progress results are visualized in the study area’s map, using a geographic information system. The goal of the geospatial visualization is show to the citizens the places where traffic events probably would occur, giving them the opportunity to change their routes avoiding traffic problems. One of the main characteristics of this approach is its use of volunteered geographic information.