CS700:Graduate Seminar in Computer Science & Informatics

Predictive Modeling of Online Human Activities

Abstract: The explosive growth and massive availability of user generated data in online social media raises tremendous challenges and opportunities to analytic study of human behavior. Effective modeling of online human behavior is fundamentally important to improve our understanding about humanities and also of practical value to the online industry. In this presentation, I will briefly introduce predictive models and algorithms we have developed that help make sense of the large-scale human behavioral data and discover knowledge to bridge the gap between theories in humanities and the practices in industry. In particular, I will describe: 1. Collaborative-competitive filtering, a novel game-theoretic framework for modeling, simulating and predicting users' online decision making behavior. 2. A series of models and algorithms for characterizing social contagion and the interplay between social relation and individual decision making.

Bio: Shuang-Hong ("Shuang") Yang is a Ph.D candidate at College of Computing, Georgia Institute of Technology. His research interests are at the interface of machine learning and social computing, focusing on developing models, algorithms and tools to make sense of the big data in online social media. He earned his B.S. in Electrical engineering from Wuhan University and M.S. in Computer Science from Chinese Academy of Sciences. He is the recipient of the SIGIR-2011 Best Student Paper Award and the PAKDD-2008 Best Student Paper Award, the nominee of the UAI-2010 Best Student Paper Award, the winnder of the Yahoo! 2011 Key Scientific Challenge, and the finalist of Facebook 2011 PhD fellow.