On Mining Anomalous Patterns in Road Traffic Streams

  • Linsey Xiaolin Pang ,
  • Sanjay Chawla ,
  • Wei Liu ,
  • Yu Zheng

International Conference on Advanced Data Mining and Applications |

Published by ADMA 2011

Publication

Large number of taxicabs in major metropolitan cities are now equipped with a GPS device. Since taxis are on the road nearly twenty four hours a day (with drivers changing shifts), they can now act as reliable sensors to monitor the behavior of traffic. In this paper we use GPS data from taxis to monitor the emergence of unexpected behavior in the Beijing metropolitan area. We adapt likelihood ratio tests (LRT) which have previously been mostly used in epidemiological studies to describe traffic patterns. To the best of our knowledge the use of LRT in traffic domain is not only novel but results in accurate and rapid detection of anomalous behavior.