a personal digital assistant, is a promising solution to
the problem of finding useful information among a sea of
texts and other web documents. By accompanying users as
they browse the Internet, the WebMate agent:
Provides URL recommendations based on a continuously
updated user profile;
Offers ever more relevant web documents based on the
"Trigger Pairs Model" approach to keyword refinement;
Responds to user feedback by selecting features from
documents the user finds relevant and incorporating
these features into the context of new queries;
Compiles a daily personal newspaper with links to documents
of interest to the user.
WebMate architecture consists of a stand-alone proxy that
monitors the user's actions to provide information for learning
and search refinement, and an applet controller that interacts
with the user. Click the icon below for a graphic representation
of the WebMate architecture:
stand-alone component is an HTTP proxy that sits between
a user's web browser and the Internet. All HTTP transactions
pass through the WebMate agent, where they can be monitored
and used as aids in agent learning. The applet controller
is the interface between the user and the stand-alone proxy.
Through it, the user expresses personal browsing preferences
and provides relevance feedback to the WebMate agent.
Institute Project Page
the WebMate website.
more information on the WebMate personal agent, see the
Chen, L. and Sycara, K., "WebMate: A Personal Agent
for Browsing and Searching." Proceedings of the
2nd International Conference on Autonomous Agents and
Multi Agent Systems, AGENTS '98, ACM, May 1998, pp.
132 - 139. [Abstract]
Download: in pdf,
Decker, K. Sycara, A. Pannu, and M. Williamson. Designing Behaviors for Information Agents. In
Proceedings of the First International Conference on
Autonomous Agents (AGENTS-97), Feb-97. (in
Sycara, K. Decker, A. Pannu, M. Williamson, and D. Zeng.
Agents. IEEE Expert, Dec-96. (in