Visit ShanghaiTech University | 中文 | How to find us
HOME > News and Events > Events
Human Mobility Synchronization and Trip Purpose Detection with Mixture of Hawkes Processes
Date: 2017/6/5             Browse: 330

Speaker:  Yanjie Fu 

Time:       June 5, 16:00-17:00.

Location: Room 1A-200, SIST Building

Inviter:    Prof.Shenghua Gao


While the semantic exploration of human mobility can benefit many applications such as smart transportation, city planning, and urban economics, there are two key questions that need to be answered: (i) What is the nature of the spatial diffusion of human mobility across functional regions? (ii) How to spot and trace the trip purposes of trajectories? To answer these questions,  we study large-scale and city-wide taxi trajectories, which are used to identify a very important property of human mobility. This property is that of synchronization. In other words, if two regions share similar spatial configurations and urban functions in a particular time period, the two regions are likely to show arrival events with similar arrival rates and trip purposes. Indeed, the synchronization property of human mobility links mobility arrivals, functional regions, and trip purposes. In addition, the purpose of a trip can be reflected by the POI topics of its origin and destination regions. Along this line, we show that the two challenges can be addressed simultaneously by exploiting human mobility synchronization and regional POI topics. To this end, we provide a unique perspective of modeling human mobility data as a stochastic process. We develop a unified probabilistic model by incorporating the impact of human mobility synchronization and POI topics of origin and destination regions into a mixture of Hawkes point processes. The proposed model is capable of modeling mobility arrivals and spotting trip purposes. In addition, we provide an effective learning algorithm for the optimization problem. Finally, we conduct intensive evaluations with a variety of real-world data and experimental results demonstrating the effectiveness of our proposed modeling method.


Dr. Yanjie Fu received his Ph.D. degree in Information Technology from Rutgers University in 2016, the B.E. degree in Computer Science from University of Science and Technology of China in 2008, and the M.E. degree in Computer Engineering from Chinese Academy of Sciences in 2011. He is currently an Assistant Professor at the University of Missouri Rolla. His research interests include data mining and big data analytics. He has research experience in industry research labs, such as Microsoft Research Asia, Huawei Research Labs, and IBM Thomas J. Watson Research Center. He has published prolifically in refereed journals and conference proceedings, such as IEEE TKDE, ACM TKDD, IEEE TMC, ACM SIGKDD, IEEE ICDM, and SIAM SDM.



SIST-Seminar 17022