Title: Discriminative Learning for Expert Search
Dr. Luo Si, Purdue University
Date: March 18, 2011
Time: 3:00 p.m.
Room: 3105 Engineering Building
Working in the information age, the most important thing may not be what you know, but who you know. In many large academic, commercial, and government organizations, it is often a crucial task to identify experts in specific topics. In real world applications of expert search, it is important to identify experts from heterogeneous types of sources, which contain documents with multiple types of associations with expert candidates. Rapid progress on expert search has been made in modeling and evaluation since the launch of TREC Enterprise Track in 2005. Most existing approaches are generative probabilistic models, which describe how a user query is generated from supporting documents of an expert candidate. On the other side, discriminative learning models have been almost absent in expert search, but discriminative models have several major advantages compared to generative models, including more accurate results and the ability to incorporate different types of features. This talk will describe two of our recent research in this area: 1). A unified model that integrates document evidence and document-candidate associations for expert search; and 2). A mixture discriminative model approach for adaptively ranking experts in heterogeneous information sources for different types of expert candidates and different types of user queries.
Dr. Si is an assistant professor in the Computer Science Department and the Statistics Department (by courtesy) at Purdue University. Dr. Si's research interests include information retrieval, machine learning techniques and applications, and text mining techniques for different applications, which have resulted in more than 60 publications. Dr. Si is an associate editor of ACM Transactions on Information System and an editorial board member of Information Processing and Management. Dr. Si received the NSF Career Award in 2008. He earned his Ph.D. degree from Carnegie Mellon University in 2006.