Winner! Provost MQP Award for Computer Science, 1998-1999
Many online newspapers provide news without consideration for reader's tastes. Collaborative filtering solves this problem by recommending articles based upon ratings by others. Collaborative filtering can be enhanced by content-based filtering when ratings are sparse. We applied a blend of collaborative and content-based filtering to an online newspaper, the Worcester Telegram and Gazette Online. Our system consists of a web-based interface, a back-end recommendation engine, and a database. We analyze results of the project and propose directions for future work.
Mark Claypool, Anuja Gokhale, Tim Miranda, Pavel Murnikov, Dmitry Netes and Matthew Sartin, Combining Content-Based and Collaborative Filters in an Online Newspaper, ACM SIGIR Workshop on Recommender Systems, Berkeley, CA, August 19, 1999.