Clustering Daily Metro Origin-Destination Matrix in Shenzhen China

Abstract:

The development of information technology gives rise to explosive growth of the amount of data. As a result, a more effective data mining method in pattern recognition is called into existence, which can properly reflect the inherent daily activity structure of metro travelers. This study is aimed to enrich the traditional clustering methods and provide practical information in dealing with traffic volume variation to the metro system operations. In this study, daily metro origin-destination (OD) data come from smart card records of Shenzhen, China, which cover 290 days and 118 stations. Principal component analysis (PCA) and singular value decomposition (SVD) are applied to conduct dimensionality reduction. Affinity propagation is then chosen to cluster the dimensionality reduced matrix to identify demand patterns of the metro OD matrix. Eleven representative categories are clustered and shown.

• Periodical: Applied Mechanics and Materials (Volume 743)
• Pages: 422-432
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.743.422
• Online since: March 2015
• Authors: Chao Yang, Fen Fan Yan*, Xiang Dong Xu
• Keywords: Affinity Propagation, Cluster Analysis, Metro OD Matrix, PCA

References

[1] Munizaga, Marcela A., and Carolina Palma. Estimation of a disaggregate multimodal public transport Origin–Destination matrix from passive smartcard data from Santiago, Chile., Transportation Research Part C: Emerging Technologies 24 (2012): 9-18.

DOI: 10.1016/j.trc.2012.01.007

[2] Chrobok, Roland, et al. Different methods of traffic forecast based on real data., European Journal of Operational Research 155. 3 (2004): 558-568.

DOI: 10.1016/j.ejor.2003.08.005

[3] Weijermars, Wendy, and Eric van Berkum. Analyzing highway flow patterns using cluster analysis., Intelligent Transportation Systems, 2005. Proceedings. 2005 IEEE. IEEE, (2005).

DOI: 10.1109/itsc.2005.1520157

[4] Friedrich, Markus, et al. Generating origin-destination matrices from mobile phone trajectories., Transportation Research Record: Journal of the Transportation Research Board 2196. 1 (2010): 93-101.

DOI: 10.3141/2196-10

[5] Roweis, Sam T., and Lawrence K. Saul. Nonlinear dimensionality reduction by locally linear embedding., Science 290. 5500 (2000): 2323-2326.

DOI: 10.1126/science.290.5500.2323

[6] Yu, Hua, and Jie Yang. A direct LDA algorithm for high-dimensional data—with application to face recognition., Pattern recognition 34. 10 (2001): 2067-(2070).

DOI: 10.1016/s0031-3203(00)00162-x

[7] Punj, Girish, and David W. Stewart. Cluster analysis in marketing research: review and suggestions for application., Journal of marketing research(1983): 134-148.

DOI: 10.1177/002224378302000204

[8] Dueck, Delbert, and Brendan J. Frey. Non-metric affinity propagation for unsupervised image categorization., Computer Vision, 2007. ICCV 2007. IEEE 11th International Conference on. IEEE, (2007).

DOI: 10.1109/iccv.2007.4408853

[9] Jackson, J. Edward. A user's guide to principal components. Vol. 587. John Wiley & Sons, (2005).

[10] Yang, Jian, et al. Two-dimensional PCA: a new approach to appearance-based face representation and recognition., Pattern Analysis and Machine Intelligence, IEEE Transactions on 26. 1 (2004): 131-137.

DOI: 10.1109/tpami.2004.1261097

[11] Frey, Brendan J., and Delbert Dueck. Clustering by passing messages between data points., science 315. 5814 (2007): 972-976.

DOI: 10.1126/science.1136800

[12] Xia, Ding-yin, et al. Local and global approaches of affinity propagation clustering for large scale data., Journal of Zhejiang University SCIENCE A9. 10 (2008): 1373-1381.

DOI: 10.1631/jzus.a0720058