Item Selection Strategies for Collaborative Filtering

Automated collaborative filtering (ACF) methods leverage the ratings-based profiles of users that are similar to some target user in order to proactively select relevant items, or predictively rate specific items, for the target user. Many of the advantages of ACF methods are derived from its contentfree approach to recommendation; it is not necessary to rely on content-based descriptions of the recommendable items, only their ratings distribution across the population of raters. Furthermore, ACF methods have an element of serendipity associated with them, as users can find items for which they would never have explicitly searched but nonetheless find interesting. The raters, items and ratings form an sparse matrix, R and the classical ACF prediction task is to predict the rating that user t wil l give to item i given that Rltl is currently empty. For example, Resnick's well-known algorithm [Resnick et al, 1994] predicts Rt,i based on t's average rating, and the rating each rater r gives to i, relative to r's average rating, and weighted by the correlation between the shared ratings of t and r.