Feature Selection in Conditional Random Fields for Map Matching of GPS Trajectories

  • Jian YangEmail author
  • Liqiu Meng
Part of the Lecture Notes in Geoinformation and Cartography book series (LNGC)


Map matching of the GPS trajectory serves the purpose of recovering the original route on a road network from a sequence of noisy GPS observations. It is a fundamental technique to many Location Based Services. However, map matching of a low sampling rate on urban road network is still a challenging task. In this paper, the characteristics of Conditional Random Fields with regard to inducing many contextual features and feature selection are explored for the map matching of the GPS trajectories at a low sampling rate. Experiments on a taxi trajectory dataset show that our method may achieve competitive results along with the success of reducing model complexity for computation-limited applications.


Map matching GPS trajectory Conditional random fields Feature selection 



We would like to thank Dr.-Ing Hongchao Fan and Prof. Chun Liu for sharing with us the Shanghai Taxi FCD dataset, and to Oliver Maksymiuk for the helpful discussion. The first author is supported by China Scholarship Council (CSC).


  1. Goh C, Dauwels J, Mitrovic N (2012) Online map-matching based on Hidden Markov model for real-time traffic sensing applications. In: ITSC 12’Google Scholar
  2. Hummel B (2006) Map matching for vehicle guidance. In: Drummond J, Billen R (eds) Dynamic and mobile GIS: investigating space and time. CRC Press, FloridaGoogle Scholar
  3. Hunter T, Abbeel P, Bayen A (2013) The path inference filter: model-based low-latency map matching of probe vehicle data. Algorithmic foundations of robotics X. Springer, Berlin, pp 591–607Google Scholar
  4. Krumm J, Letchner J, Horvitz E (2007) Map matching with travel time constraints. In: SAE world congressGoogle Scholar
  5. Lafferty J, McCallum A, Pereira F (2001) Conditional random fields: probabilistic models for segmenting and labeling sequence data. In: ICML 2001 pp 282–289Google Scholar
  6. Li Y, Huang Q, Kerber M (2013) Large-scale joint map matching of GPS traces. In: ACM GIS’13, pp 1–10Google Scholar
  7. Lou Y, Zhang C, Zheng Y, Xie X, Wang W, Huang Y (2009) Map-matching for low-sampling-rate GPS trajectories. In: ACM GIS’09. ACM Press, p 352Google Scholar
  8. Newson P, Krumm J (2009) Hidden Markov map matching through noise and sparseness. In: Proceedings of the 17th ACM GIS ’09. p 336Google Scholar
  9. Ng AY (1998) On feature selection: learning with exponentially many irrevelant features as training examples. In: ICML’98. pp 404–412Google Scholar
  10. Quddus M, Ochieng W, Noland R (2007) Current map-matching algorithms for transport applications: state-of-the art and future research directions. Transp Res Part C: Emerg Technol 15(5):312–328CrossRefGoogle Scholar
  11. Schmidt M (2010) Graphical model structure learning with L1-regularization. University of British Columbia, CanadaGoogle Scholar
  12. Schmidt M, Fung G, Rosaless R (2009) Optimization methods for L1-regularization. University of British Columbia, CanadaGoogle Scholar
  13. Sha F, Pereira F, Science I (2003) Shallow parsing with conditional random fields. In: Proceedings of ACL’03, vol 1, pp 134–141, ACLGoogle Scholar
  14. Sutton C (2012) An introduction to conditional random fields. Found Trends® Mach Learn 4(4):267–373CrossRefGoogle Scholar
  15. Yuan J, Zheng Y, Zhang C, Xie X, Sun G-Z (2010) An interactive-voting based map matching algorithm. In: 2010 Eleventh international conference on mobile data management pp 43–52Google Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Lehrstuhl für KartographieTechnische Universität MünchenMunichGermany

Personalised recommendations