Distributing Information for Collaborative Filtering on Usenet Net News.

As part of the Information Revolution,” the amount of raw information available to computer users has increased as never before. Unfortunately , there has been a corresponding jump in the amount of unrelated information users must search through in order find information of interest. Harnessing the power of multiple users to form a collaborative fillter provides a robust wayy of helping direct users to the information that will be most useful to them. To test this idea, we have designed a large scale collaborative filltering system tuned to help users extract information from Usenet Net News. In this thesis we demonstrate a system for collaborative filltering that can scale up to encompass the large distributed information sources of which Usenet Net News is an example. Our system provides varying levels of anonymity to protect the interests of the users, as w ell as means of minimizing the load placed on the existing information source. We believe the system will be especially good at three tasks: supporting users as they explore new areas of interest; providing users a way of keeping up to date on areas they already have familiarity with; and at providing extra information to other fillters.