A Evaluation Model for Valid Prediction of Cabs’ Number Change from Urban Cabs Trajectories

Abstract:

the large scale urban cabs trajectories are used for prediction of cabs number change in some applications such as knowledge discovery and abnormal traffic alarm. The traditional methods calculate for many invalid spaces consume much time on invalid calculation. After observing many datasets with the great number of cabs trajectories points, we find a rule that the periodicity and stabilization of cabs group number change will be clearer as the number of cabs trajectories points increasing in some space, and the prediction accurate of cabs number change is also have relationship with the periodicity and stabilization. To validate this rule, we first model the periodicity and stabilization of cabs group number change. And a function is found to well fit the model value as the change of cabs group number, proving the rule we have found. Finally, we show that how this rule can be used to decrease the calculation time of valid traffic prediction. Experiment shows that the rule we discovered exists in Beijing and San Francisco datasets and the rule can conveniently and efficiently used to valid traffic prediction.

• Periodical: Applied Mechanics and Materials (Volume 596)
• Pages: 259-262
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.596.259
• Online since: July 2014
• Authors: Wu Bin Ma*, Su Deng, Hong Bin Huang
• Keywords: Evaluation Model, Location Based Service, Prediciton

References

[1] V. W. Zheng, et al., Towards mobile intelligence: Learning from GPS history data for collaborative recommendation, Artificial Intelligence, (2012).

DOI: 10.1016/j.artint.2012.02.002

[2] J. Yuan, et al., Discovering Regions of Different Functions in a City Using Human Mobility and POIs, in KDD, (2012).

[3] J. Yuan, et al., Driving with Knowledge from the Physical World, in KDD11, (2011).

[4] J. Yuan, et al., T-Drive: Enhancing Driving Directions with Taxi Drivers' Intelligence, ACM SIGSPATIAL 2010, (2012).

[5] S. Ma, et al., T-Share: A Large-Scale Dynamic Taxi Ridesharing Service, in ICDE13, (2013).

[6] X. LI, et al., Prediction of urban human mobility using large-scale taxi traces and its applications, Front. Comput. Sci., vol. 6, pp.111-121, (2012).

[7] W. Mathew, et al., Predicting Future Locations with Hidden Markov Models, in Ubicom, (2012).

[8] PengC., et al., Collective Human Mobility Pattern from Taxi Trips in Urban Area, PLoS ONE, vol. 7, (2012).

[9] Available: http: /cabspotting. org.

[10] Y. Zheng, et al., Mining Interesting Locations and Travel Sequences from GPS Trajectories, in WWW, Madrid, Spain, (2009).

[11] S. G. Mallat, A theory for multiresolution signal decomposition: the wavelet representation, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, pp.674-693, (1989).

DOI: 10.1109/34.192463

[12] X. Liang, et al., The scaling of human mobility by taxis is exponential, Physica A, (2011).

[13] W. Liu, et al., Discovering Spatio-Temporal Causal Interactions in Traffic Data Streams, in KDD, San Diego, California, USA, (2011).

[14] X. J. Ban and M. Gruteser, Towards Fine-Grained Urban Traffic Knowledge Extraction Using Mobile Sensing, in UrbComp, (2012).

DOI: 10.1145/2346496.2346514

[15] K. Zheng, et al., On Discovery of Gathering Patterns from Trajectories, in ICDE (2013).

[16] P. S. Castro, et al., Urban traffic modelling and prediction using large scale taxi GPS traces, (2012).

[17] N. J. Yuan, et al., T-Finder: A Recommender System for Finding Passengers and Vacant Taxis, in UbiComp 2011, (2011).