Title: User Proling and Privacy Preserving from Multiple Social networks

Time: 8:30, June 16, 2016

Venue: room 202, Office Building

Abstract: 

User profiling, which aims to infer users' unobservable information based on observable information such as individual's behaviour or utterances, is the basis for many applications, such as personalized recommendation, and expert finding. Traditional user profiling conducted with traditional medium, such as document records, is always hindered by the limited data sources. Recent years, the proliferation of social media has opened new opportunities for user profiling. Moreover, as different social networks provide different services, increasing number of people are involved in multiple social networks. Different aspects can be revealed by different social networks. Therefore, to comprehensively learn users' profiles, it is time to shift from a single social network to multiple social networks. Therefore, this talk aims to investigating user profiling across multiple social networks. In particular, it covers the studies in two general scenarios in user profiling across multiple social networks, where a single task and multiple tasks are involved respectively. Meanwhile, noticed the potential of social media in user profiling would put users at high privacy risks, it also proposed a scheme for privacy preserving from social media.

This talk will first introduce a novel scheme for multi-source mono-task learning to infer users' attributes, such as volunteerism tendency, which involves a single task. The proposed scheme is able to model both the source confidence and source consistency simultaneously. Then a multi-source multi-task learning scheme to infer users attributes, such as interest, which usually involves multiple related tasks, will be presented. The proposed scheme jointly regularizes two important aspects: source consistency and task relatedness. Finally,  this talk will show a scheme for privacy preserving to reduce users' privacy risks on social media. In particular, it proposes a taxonomy to comprehensively characterize users' personal aspects. With the guidance of such taxonomy, we correspondingly propose a multi-task learning scheme to identify the potential privacy leakage.

Bio: 

Miss Xuemeng Song is currently a fourth-year Ph.D. student from the School of Computing, National University of Singapore. She received her B.E. degree from the Department of Electronic Engineering & Information Science, University of Science and Technology of China in 2012. Her research interests are information retrieval and social network analysis. She has published several papers in the top venues, such as SIGIR, IJCAI, and TOIS. In addition, she has served as reviewers for many top conferences and journals, such as TKDD, TMM, ICMR and MMM.