Analyzing Traffic Accidents based on the Integration of Official and Crowdsourced Data

Authors

  • Salatiel Ribeiro dos Santos Universidade Federal de Minas Gerais
  • Clodoveu Augusto Davis Jr. Universidade Federal de Minas Gerais
  • Rodrigo Smarzaro Universidade Federal de Minas Gerais

DOI:

https://doi.org/10.5753/jidm.2017.1607

Keywords:

accidents, Belo Horizonte, spatial data integration, waze

Abstract

Geographic information of public interest is routinely produced by several public agencies. At the same time, the use of smartphones and other mobile devices generates an increasing amount of unofficial georeferenced data. Although official data have usually higher reliability, it takes longer for governmental organizations to put together relevant datasets and make them available, while the opposite occurs with unofficial data. This work explores the potential for integrating data from official and unofficial sources, as part of a project that aims to verify possible roles for unofficial or crowdsourced data, as replacements or to complement official sources. The article presents a case study with two traffic accident datasets in the city of Belo Horizonte, Brazil. We compare official traffic accident data to unofficial data collected from Waze, a mobile app dedicated to helping users fight traffic congestion. We found that seven percent of accidents reported by official sources have also been reported by users of Waze. Accidents reported only by official sources are concentrated in the central region, while those recorded by Waze are mostly on some major roads all over the city. An analysis on the possible influence of weather is also presented, as well as the identification of accident hotspots from the integrated dataset.

Downloads

Download data is not yet available.

Downloads

Published

2017-09-27

How to Cite

Santos, S. R. dos, Davis Jr., C. A., & Smarzaro, R. (2017). Analyzing Traffic Accidents based on the Integration of Official and Crowdsourced Data. Journal of Information and Data Management, 8(1), 67. https://doi.org/10.5753/jidm.2017.1607

Issue

Section

GeoInfo 2016