Understanding Taxi Driving Behaviors from Movement Data
Understanding taxi mobility has significant social and economic impacts on the urban areas. The goal of this paper is to visualize and analyze the spatiotemporal driving patterns for two income-level groups, i.e. high-income and low-income taxis, when they are not occupied. Specifically, we differentiate the cruising and stationary states of non-occupied taxis and focus on the analysis of the mobility patterns of these two states. This work introduces an approach to detect the stationary spots from a large amount of non-occupied trajectory data. The visualization and analysis procedure comprises of mainly the visual analysis of the cruising trips and the stationary spots by integrating data mining and visualization techniques. Temporal patterns of the cruising trips and stationary spots of the two groups are compared based on the line charts and time graphs. A density-based spatial clustering approach is applied to cluster and aggregate the stationary spots. A variety of visualization methods, e.g. map, pie charts, and space-time cube views, are used to show the spatial and temporal distribution of the cruising centers and the clustered and aggregated stationary spots. The floating car data collected from about 2000 taxis in 47 days in Shanghai, China, is taken as the test dataset. The visual analytic results demonstrate that there are distinctive cruising and stationary driving behaviors between the high-income and low-income taxi groups.
KeywordsTaxi driving behavior Mobility pattern Movement data
This work is partially funded by the China Scholarship Council. We would like to thank Prof. Chun Liu (School of Surveying and Geoinformatics, Tongji University) for sharing with us the Shanghai Taxi FCD dataset.
- Ester, M., Kriegel, H., Sander, J., & Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. In E. Simoudis, J. Han, U. Fayyad & M. Usama (Eds.), Proceedings of the Second International Conference on Knowledge Discovery and Data Mining (KDD-96) (pp. 226–231, ISBN 1–57735-004-9). Palo Alto: AAAI Press.Google Scholar
- Guo, H., Wang, Z., Yu, B., Zhao, H., & Yuan, X. (2011). TripVista: triple perspective visual trajectory analytics and its application on microscopic traffic data at a road intersection. In Proceedings of IEEE Pacific Visualization Symposium (PacificVis 2011) (pp. 163–170), Hong Kong, March 1–4, 2011.Google Scholar
- Huber, W., Lädke, M., & Ogger, R. (1999). Extended floating-car data for the acquisition of traffic information. In Proceedings of the 6th World Congress on Intelligent Transport Systems.Google Scholar
- Liu, L., Andris, C., Biderman, A., & Ratti, C. (2009). Uncovering taxi driver’s mobility intelligence through his trace. IEEE Pervasive Computing.Google Scholar
- Tominski, C., Schumann, H., Andrienko, G., & Andrienko, N. (2012). Stacking-based visualization of trajectory attribute data. IEEE Transactions on Visualization and Computer Graphics, 18(12), 2012.Google Scholar
- Yuan, J., Zheng, Y., Zhang, L., & Xie, X. (2012a), T-finder: A recommender system for finding passengers and vacant taxis. IEEE Transactions on Knowledge and Data Engineering (TKDE).Google Scholar
- Zheng, Y., & Zhou, X. (2011). Computing with spatial trajectories. Berlin: Springer. ISBN:978-1-4614-1628-9.Google Scholar