PoI as the Physical Knowledge for Human Behavior Prediction

Talk Abstraction

Human behavior prediction, including both the physical-world mobility and activities and cyber space web browsing, are critically important for ubiquitous computing system. The performance of prediction greatly depends on the quantity and `quality' of data. In real-world systems, collecting such data reflecting human behavior can be costly or impossible due to practical limitations. We address this problem by leveraging a key insight – people’s intention behind their behavior is highly correlated with the nearby point of interests (PoI), which are the crowd sourcing labels accounts for the function of city regions and intentions of human activities. In this talk, we first present the first population-level, city-scale prediction of application usage on smartphones by developing a technique that leverages transfer learning to predict which applications are most popular and estimate the whole usage distribution based on the POI information of that particular location. Then, we introduce how to using POI as the knowledge to predict human mobility behaviors from both the individual and crowd level. Our findings pave the way for predicting which apps are relevant to users, and how they move given their current location with the physical knowledge of POI.

Speaker Bio

Dr. Yong Li (http://fi.ee.tsinghua.edu.cn/~liyong/) is the Associate Professor of the Department of Electronic Engineering, Tsinghua University. He received the B.S. degree from Huazhong University of Science and Technology in 2007, and the M. S. and the Ph. D. degrees in Electrical Engineering from Tsinghua University, in 2009 and 2012, respectively. His research interests are in the areas of ubiquitous computing, data mining and mobile computing.

Dr. Li has served as General Chair, TPC Chair, TPC Member for several international workshops and conferences, and he is on the editorial board of two IEEE journals. His papers published on UbiComp, KDD, WWW, etc. have total citations more than 6300. Among them, ten are ESI Highly Cited Papers in Computer Science, and four receive conference Best Paper (run-up) Awards. He received IEEE 2016 ComSoc Asia-Pacific Outstanding Young Researchers and Young Talent Program of China Association for Science and Technology.