• Summary
• Related Documents
  • Active Bibliography
  • Co-citation
• Version History

Abstract:

Document ® ltering is increasingly deployed in W eb environments to reduce information overload of users. W e formulate online information ® ltering as a reinforcement learning problem, i.e., TD(0). The goal is to learn user pro ® les that best represent information needs and thus maximize the expected value of user relevance feedback. A method is then presented that acquires reinforcement signals automatically by estimating user’s implicit feedback from direct observations of browsing behaviors. This ``learning by observation’ ’ approach is contrasted with conventional relevance feedback methods which require explicit user feedbacks. Field tests have been performed that involved 10 users reading a total of 18,750 HTML documents during 45 days. Compared to the existing document ® ltering techniques, the proposed learning method showed superior performance in information quality and adaptation speed to user preferences in online ® ltering. With the rapid progress of computer technology in recent years, electronic information has been explosively increased. This trend is especially remarkable on the Web. As the availability of the information increases, the need for ® nding more relevant information on the Web is growing (Belkin & Croft,

Citations