Improving Ontology-Based User Profiles

Personalized Web browsing and search hope to provide Web information that matches a user's personal interests and thus provide more effective and efficient information access. A key feature in developing successful personalized Web applications is to build user profiles that accurately represent a user's interests. The main goal of this research is to investigate techniques that implicitly build ontology-based user profiles. We build the profiles without user interaction, automatically monitoring the user's browsing habits. After building the initial profile from visited Web pages, we investigate techniques to improve the accuracy of the user profile. In particular, we focus on how quickly we can achieve profile stability, how to identify the most important concepts, the effect of depth in the concept-hierarchy on the importance of a concept, and how many levels from the hierarchy should be used to represent the user. Our major findings are that ranking the concepts in ...

• Type: Conference
• Year: 2004
• Where: RIAO
• Authors: Joana Trajkova, Susan Gauch
• Keywords: RIAO 2004, RIAO 2007, User Profile, User's Browsing Habits, User's Personal Interests