Abstract

We discuss learning a profile of user interests for recommending information sources such as Web pages or news articles. We describe the types of information available to determine whether to recommend a particular page to a particular user. This information includes the content of the page, the ratings of the user on other pages and the contents of these pages, the ratings given to that page by other users and the ratings of these other users on other pages and demographic information about users. We describe how each type of information may be used individually and then discuss an approach to combining recommendations from multiple sources. We illustrate each approach and the combined approach in the context of recommending restaurants.

Citations

2307	Learning internal representations by error propagation - Rumelhart, Hinton, et al. - 1986
1302	WordNet: an online lexical database - Miller, Beckwith, et al. - 1990
946	Agents that reduce work and information overload - Maes - 1994
875	Grouplens: An open architecture for collaborative filtering of netnews - Resnick, Iacovou, et al. - 1994
728	Relevance feedback in information retrieval - Rocchio - 1971
620	Automatic text processing - Salton - 1988
556	The weighted majority algorithm - Littlestone, Warmuth - 1994
538	Improving retrieval performance by relevance feedback - Salton, Buckley - 1990
516	Letizia: An Agent that Assists Web Browsing - Lieberman - 1995
380	Fab: Content-Based, Collaborative Recommendation - Balabanovic, Shoham - 1997
228	Learning and Revising User Profiles: The Identification of Interesting Web Sites - Pazzani, Billsus - 1997
216	Training algorithms for linear text classi ers - Lewis, Shapire, et al. - 1996
196	Exponentiated gradient versus gradient descent for linear predictors - Kivinen, Warmuth - 1997
114	Empirical support for Winnow and weighted-majority based algorithms: results on a calendar scheduling domain - Blum - 1995
110	Combining classifiers in text categorization - Larkey, Croft - 1996
104	An adaptive web page recommendation service - Balabanovic - 1997
66	Newsweeder: Learning to filter news - Lang - 1995
46	Learning in the presence of finitely or infinitely many irrelevant attributes - Blum, Hellerstein, et al. - 1995
45	Lifestyle finder - intelligent user profiling using large-scale demographic data - Krulwich - 1997