CONCEPT-BASED PERSONALIZED WEB SEARCH

PhD Thesis Proposal Defence


Title: "CONCEPT-BASED PERSONALIZED WEB SEARCH"

by

Mr. Kenneth Wai-Ting Leung


Abstract:

Personalized search is an important means to improve the retrieval 
effectiveness of a search engine. Most commercial search engines return the 
same set of results to all users who ask the same query. However, different 
users may have different preferences on the result set. In this proposal, we 
present an effective method to mine a user's conceptual preferences from search 
engine clickthrough data, and adjust the search result ranking according to the 
extracted preferences to improve the retrieval effectiveness for the user. Our 
approach employs an Ontology-based User Profile (OUP) that represents a user 
profile as an ontology. We then derive an extended set of conceptual 
preferences based on the ontology and the user's clickthroughs. The user 
profile is input to a Support Vector Machine (SVM) to learn a concept 
preference vector for adapting a personalized ranking function to re-rank the 
search results. Our initial results show that the top-10 precision achieved by 
OUP is almost three times higher compared to the method without employing 
personalization at all, and about 60% better than that of the hierarchical user 
profile proposed by Xu et al. Moreover, our method can significantly improve 
the users' average clicked ranks by about 65% compared to other methods that do 
not employ concept ontology. Based on these results, we will further extend our 
research in the areas of community-based personalization and location-based 
personalization.


Date:  			Thursday, 10 December 2009

Time:                   2:00pm - 4:00pm

Venue:                  Room 1504
 			lifts 25/26

Committee Members:      Prof. Dik-Lun Lee (Supervisor)
 			Dr. Wilfred Ng (Supervisor)
 			Dr. Raymond Wong (Chairperson)
 			Dr. Lei Chen
 			Dr. Qiong Luo


**** ALL are Welcome ****