2011 Poster Sessions : Friendship and Mobility: User Movement In Location-Based Social Networks

Student Name : Eunjoon Cho, Seth Myers
Advisor : Jure Leskovec
Research Areas: Computer Systems
Abstract:
Even though our movement and mobility patterns have high degree of freedom and variation, human movement exhibits high structural patterns due to geographic and social constraints. Humans experience a combination of strong geographically limited periodic movement combined with seemingly random jumps influenced by their social networks. Using cell phone location tracking data from a large European country as well as the data from two online location based social networks we aim to understand what basic laws govern human motion and dynamics. We investigate how periodic behavior combined with the influence from the social network structure shape human mobility. We observe strong correlation between the friendship and mobility features of a user. We show that social relationships, however, are not sufficient to explain all human movement, and thus develop a hybrid model that also captures the periodic short range movements of users. We show that our model reliably predicts future movement of a user and gives an order of magnitude better performance than standard models of human mobility.

Bios:
Eunjoon Cho is a PhD candidate in the Electrical Engineering department at Stanford University. He did his undergraduate at KAIST in Electrical Engineering and his masters in Electrical Engineering at Stanford University. His current research focus is in cognitive memory and its application to speech recognition and has previously done various research on human mobility.

Seth A. Myers is a third year PhD student in the Institute for Computational and Mathematical Engineering at Stanford University. He did his undergraduate at Northwestern University where he triple majored in mathematics, computational physics, and integrated science. His primary research focuses are social network inference and human mobility.

Jure Leskovec is an assistant professor of Computer Science at Stanford University where he is a member of the InfoLab and the AI lab. He joined the department in September 2009.

In 2008/09 Jure was a postdoctoral researcher at Cornell University working with Jon Kleinberg and Dan Huttenlocher. He completed his Ph.D. in Machine Learning Department, School of Computer Science at Carnegie Mellon University under the supervision of Christos Faloutsos in September 2008.

Jure did his undergraduate in computer science at University of Ljubljana, Slovenia. He also work with the Artificial Intelligence Laboratory, Jozef Stefan Institute, Ljubljana, Slovenia.