Unraveling traveler mobility patterns and predicting user behavior in the Shenzhen metro system
place - asia, place - urban, mode - subway/metro, ridership - behaviour, planning - methods, technology - passenger information
Smart card data, mobility patterns, entropy, hidden Markov model, Markov chain model
Over the last few years, cities have made available large volumes of smart card data that shed light on the urban dynamics of transit users. This research uses metro card data from Shenzhen, China, to recognize individual mobility patterns and predict travelers' future movements. Joint entropy is proposed to measure the regularity of spatio-temporal patterns and travelers are divided into three groups, i.e. regular users, variable users and irregular users, based on the entropy value. Revised Markov chain model and hidden Markov model (HMM) are then introduced to predict individuals' future movement. We observe that the models predict with a high level of accuracy of 84.46%, 78.79% and 73.07% for three groups in the HMM. This study shows the potential to predict travel patterns and enriches traditional pattern recognition and prediction methods for modeling urban mobility. It also helps reveal structural properties of human behavior in urban metro systems.
Permission to publish the abstract has been given by Taylor&Francis, copyright remains with them.
Yang, C., Yan, F., & Ukkusuri, S.V. (2018). Unraveling traveler mobility patterns and predicting user behavior in the Shenzhen metro system. Transportmetrica A: Transport Science, Vol. 14, pp. 576-597.