From data to knowledge: city-wide traffic flows analysis and prediction using bing maps

Abstract

Traffic jam is a common contemporary society issue in urban areas. City-wide traffic modeling, visualization, analysis, and prediction are still challenges in this context. Based on Bing Maps information, this work aims to acquire, aggregate, analyze, visualize, and predict traffic jam. Chicago area was evaluated as case study. The flow intensity (free or congested) was analyzed to allow the identification of phase transitions (shocks in the system). Also, a prediction model was developed based on logistic regression to correct discovery future flow intensities for a target street.