Experience with Learning Agents Which Manage Internet-Based Information

Machine Learning in Information Access

Volume

Papers from the 1996 AAAI Spring Symposium

Contents

To provide assisistance with tasks such as retrieving USENET news articles or identifying interesting Web pages, an intelligent agent requires information about a user’s interests and needs. Machine learning techniques are now being used to acquire this information. A general architecture is presented, and two approaches to learning through observation are described. An instautiation of the architecture is then evaluated.