Predicting Bus Ridership



Dynamics of a city, among other factors, are characterized by the traveling patterns of its dwellers. Accurate knowledge of human mobility would have applications, e.g., in urban design, the optimization of public transportation operating costs, and the improvement of public transportation services. In this paper we combine a large scale bus transportation dataset with publicly available data sources. We propose a Gaussian process based approach for modeling and predicting bus ridership. We use data collected from Lisbon, Portugal to validate our approach and demonstrate superior prediction accuracy of Gaussian process compared to a probabilistic baseline predictor.


Gaussian process, modeling, urban computing


Third International Workshop on Pervasive Urban Applications (PURBA) in conjunction with ACM International Joint Conference on Pervasive and Ubiquitous Computing, September 2013

Cited by

No citations found