Keynote Speeches

Prof. Sung-Hyon Myaeng


Director, KAIST-Microsoft Research Collaboration Center (KMCC) (2010.8 ~ present)

Professor, Department of Computer Science, Korea Advanced Institute of Science and Technology (2009. 3 - present)


Dr. Sung-Hyon Myaeng is currently a professor in Department of Computer Science and the head of Web Science & Technology Division at Korea Advanced Institute of Science and Technology (KAIST). He is also the Director of KAIST-Microsoft Research Collaboration Center (KMCC). Previously he was on the faculty at Syracuse University, USA, where he was granted tenure. He earned his MS and Ph. D. from Southern Methodist University, Texas, USA in 1985 and 1987, respectively. His current research interests are: information retrieval (esp. contextual ad searching, mobile searching, & text analysis for unconventional search criteria), text mining (opinion mining, trend analysis, relation extraction, & categorization), and context-aware computing including extraction and use of commonsense knowledge, especially in human activities. He has been on editorial boards of international journals including ACM Transactions on Asian Information Processing (TALIP) and Journal of Computer Processing of Oriental Languages (JCOPL), both as an associate editor and Information Processing and Management. He also has served on program committees of many reputable international conferences in the areas of information retrieval, natural language processing, and digital libraries, including his role as a co-program chair for ACM SIGIR, 2002 and 2008. In 2008, he won an award from Microsoft Research, based on global competition for the RFP ‘Beyond Search – Semantic Computing and Internet Economics’. He received Digital Innovation Award from Hankookilbo (Korea Daily) in 2002 for the development of a new technique called Virtual Document Digital Library System.

Keynote: Enabling Proactive Search

Wednesday, Dec. 1st, 2010, 9:00 am ~ 10:00 am

More than ever, people use search technologies ubiquitously with smart devices including smart phones and TVs. People look for not only information but also objects like people, shops, and mobile applications, and even specific events and services. With more diversified search targets, it is not sufficient anymore to rely on users’ search queries and react to them. A search system needs to behave proactively in anticipation of what’s required for the user’s task at hand. This talk will start from this premise and show our ongoing efforts to enable proactive search and action recommendations that are becoming increasingly important with smart devices.

Prof. ChengXiang Zhai


Associate Professor, Department of Computer Science, University of Illinois at Urbana-Champaign


ChengXiang Zhai is an Associate Professor of Computer Science at the University of Illinois at Urbana-Champaign, where he also holds a joint appointment at the Institute for Genomic Biology, Statistics, and the Graduate School of Library and Information Science. He received a Ph.D. in Computer Science from Nanjing University in 1990, and a Ph.D. in Language and Information Technologies from Carnegie Mellon University in 2002. He worked at Clairvoyance Corp. as a Research Scientist and a Senior Research Scientist from 1997 to 2000. His research interests include information retrieval, text mining, natural language processing, machine learning, and bioinformatics. He is an Associate Editor of ACM Transactions on Information Systems, and Information Processing and Management, and serves on the editorial board of Information Retrieval Journal. He is a program co-chair of ACM CIKM 2004 , NAACL HLT 2007, and ACM SIGIR 2009. He is an ACM Distinguished Scientist, and received the 2004 Presidential Early Career Award for Scientists and Engineers (PECASE), the ACM SIGIR 2004 Best Paper Award, an Alfred P. Sloan Research Fellowship in 2008, and an IBM Faculty Award in 2009.


Maximum Personalization: User-Centered Adaptive Information Retrieval

Thursday, Dec. 2nd, 2010, 8:35 am ~ 9:35 am

Precise understanding of a user's information need is critical for a search engine to deliver optimal search results to users. Unfortunately, the current search engines do not really know individual users well, limiting their ability to optimize the search results for each individual user or help users in case of long-tail difficult queries.

In this talk, I will present a new retrieval strategy called user-centered adaptive information retrieval (UCAIR) which would break this limitation and achieve maximum personalization. With UCAIR, each user would "own" a personalized intelligent search agent that would know the user very well through observing the user's information seeking behavior from the client-side, and would exploit its "complete" knowledge about the user to interactively optimize the search results for a particular user.

As a theoretical foundation for UCAIR, I will present a general decision-theoretic framework for optimal interactive information retrieval, in which the system iteratively responds to each user action by choosing an optimal system action based on all the available user information and search context. I will present several algorithms for personalizing search results, and use them to show that such a general decision-theoretic view of retrieval can naturally suggest many interesting ways to achieve personalization in an interactive search system, and that UCAIR can significantly improve the utility of the current document-centered search engines by maximally optimizing search results for each individual user.