User Interest Detection on Web Pages for Building Personalized Information Agent

Yin Liu, Wenyin Liu, Changjun Jiang

Abstract: In this paper we propose a vision-based approach to detecting user’s interests on web pages. This approach first splits a web page into several blocks of topics according to visual cues and then analyzes the layout structure of the page based on these blocks. The user’s interests are detected according to his browsing behaviors on the blocks and the layout structure of the web page. A personalized information agent is also developed to automatically extract new content from those web pages according to the user’s interests. Our experiments show that our proposed approach to detecting user’s interests on web pages achieves satisfactory accuracy and the personalized information agent significantly saves user’s browsing time and improves user’s browsing experience.