Speaker: Guandong Xu Time: 20 Dec., 2012 13:30-14:30 Venue: Room 305, Computer Building, Zhangjiang Campus Contact: 王新 xinw@fudan.edu.cn Introdcution of the talk:Web advertising, a form of online advertising which uses the Internet as an intermediate to post product or service information and attract customers, has become one of most important marketing channels. As one prevalent type of Web advertising, contextual advertising refers to “finding the ‘best match’ between a user in a given context and a suitable advertisement”, to provide a better user experience and as a result increase the user’s ad-click rate. However, due to the intrinsic problems of homonymy and polysemy, the low intersection of keywords, and a lack of sufficient semantics, traditional keyword matching techniques are not able to effectively handle contextual matching and retrieve relevant ads for the user, resulting in an unsatisfactory performance in ad selection.In this talk, we introduce a new contextual advertising approach to overcome these problems, which uses Wikipedia thesaurus knowledge to enrich the semantic and lexical expression of a targeted page (or a candidate ad). First, we map each page (or ad) into a keyword vector, upon which two additional feature vectors, i.e., the Wikipedia concept and category vector derived from the Wikipedia thesaurus structure, are then constructed. Second, in order to determine the relevant ads for a given page, we propose a linear similarity fusion mechanism, which combines the above three feature vectors in a unified manner. Last, we validate our approach by using a set of real ads, real pages along with the external Wikipedia thesaurus containing more than 730,000 concepts and 24,000 categories. The experimental results show that our approach outperforms the conventional keyword matching and Wikipedia matching approaches, and can substantially improve the performance of ad selection. Bio of the speaker：Dr. Guandong Xu has received his PhD degree in Computer Science from Victoria University. He is now a lecturer in the Advanced Analytics Institute at University of Technology, Sydney. Prior to this, he worked as Research Fellow and Postdoc in the Centre for Applied Informatics at Victoria University, Australia, Aalborg University, Denmark and University of Tokyo, Japan. His research interests cover Data management and Analytics, Data Mining, Machine Learning; Information retrieval and processing, Web search; Intelligent Web Systems, Web mining, Web Communities, Web Personalization and Recommender Systems; Social Network Analysis, Social Media Mining; as well as Social Informatics and Health Informatics. He has extensively had 60+ publications in referred international journals and conferences proceedings. He is the lead author of two scientific books published by Springer and IGI publisher. He is active in organizing or serving for program committees for a number of international conferences and workshops. He is the assistant Editor-in-Chief for World Wide Web Journal, and the guest editor of special issue with World Wide Web Journal, the Computer Journal, Journal of Software and Systems and International Journal of Social Network Mining on the related topics of social network analysis, social web mining, and social knowledge discovery and management.
