WebWatcher is a "tour guide" agent for the world wide web.
Once you tell it what kind of information you seek, it accompanies you
from page to page as you browse the web, highlighting hyperlinks that
it believes will be of interest. Its strategy for giving advice is
learned from feedback from earlier tours.
WebWatcher Instances
WebWatcher can help you search for information starting from any of
the following pages, but it has learned the most for the first of
these. Currently WebWatcher is online only on an irregular basis. You
might want to take a look at the following demo instead.
WebWatcher gives tours to many people (over 8,500 thus far), and
learns to become a specialist at a particular web site. In contrast,
our
Personal WebWatcher project stays with a single user, becoming a
specialist in that user's interests.
Publications
WebWatcher: A Tour Guide for the
World Wide Web , T. Joachims, D. Freitag, T. Mitchell,
Proceedings of IJCAI97, August 1997 (longer
version internal CMU technical report September 1996). Abstract: We describe WebWatcher as a tour guide agent for the
web, the learning algorithms used by WebWatcher, experimental results
based on learning from thousands of users, and lessons learned from
this case study of tour guide agents.
WebWatcher: Machine Learning and Hypertext, Fachgruppentreffen Maschinelles Lernen, Dortmund, Germany, August 1995. Abstract: A description of how hypertext structure can be used to cluster WWW pages.
Welcome to the WebWatcher Project
