Item-based Collaborative Filtering Recommendation Algorithms-英文文献.pdf

Item-Based Collaborative Filtering Recommendation Algorithms Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl fsarwar, karypis, konstan, riedl g@ GroupLens Research Group/Army HPC Research Center Department of Computer Science and Engineering University of Minnesota, Minneapolis, MN 55455 ABSTRACT through all the available information to nd that which is most valuable to us. Recommender systems apply knowledge discovery techniques to the problem of making p ersonalized recommendations for One of the most promising such technologies is col labora- information, pro ducts or services during a live interaction. tive ltering [19, 27, 14, 16]. Collab orative ltering works by These systems, esp ecially the k-nearest neighb or collab ora- building a database of preferences for items by users. A new tive ltering based ones, are achieving widespread success on user, Neo, is matched against the database to discover neigh- the Web. The tremendous growth in the amount of avail- bors, which are other users who have historically had similar able information and the numb er of visitors to Web sites in taste to Neo. Items that the neighb ors like are then recom- recent years p oses some key challenges for recommender sys- mended to Neo, as he will probably also like them. Collab- tems. These are: pro ducing high quality recommendations, orative ltering has b een very successful in b oth research p erforming many recommendations p er second for millions and practice, and i