---------------------------------------- ----------------------------------------

CMU Tilted Square Logo Personal WebWatcher, Project Page

---------------------------------------- ----------------------------------------

Personal WebWatcher

Overview

Personal WebWatcher is a "personal" agent that 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. Personal WebWatcher is mainly inspired by WebWatcher.

Unlike WebWatcher, Personal WebWatcher is structured to specialize for a particular user, modeling her/his interests. It ``watches over the users shoulder'' the similar way WebWatcher does, but it avoids involving the user in its learning process (it doesn't ask the user for any keywords or opinions about pages). It solely records the addresses of pages requested by the user and highlights interesting hyperlinks. In the learning phase (typically during the night), requested pages are analyzed and a model of user interests is generated/updated. This model is used to give advice for hyperlinks on retrieved HTML-pages requested by and presented to the user via Web browser.

More about Personal WebWatcher

  • Feature selection
  • Machine Learning for better Web browsing
  • Machine Learning

    Project Members

    This project is a part of CMU Text Learning Group work and strongly related to the Yahoo Planet project, the Word-Mining project and the PhD thesis project: Machine Learning on non-homogeneous, distributed text data ----------------------------------------

    Dunja Mladenic
    Last modified: Fri Jun 29 20:08:02 EDT 2001