There is a lot of interest in recommending items to a user. Suppose a user goes to an e-commerce site, what should be recommended to the user? Items might be recommended based upon what a user previously liked, bought, or what their friends liked. Recommenders deal with discovering new items for which a user could have a higher preference.
Recommender systems typically produce a list of recommendations in one of two ways—through collaborative or content-based filtering. Collaborative filtering based approaches build a model for recommendation from a user's past behavior; for example, based on items previously purchased or selected by the user, based on ratings given to items previously purchased or selected by the user, based on decisions made by users similar to the current user. The model built using past behavior can then be used to recommend items to the user. Content-based filtering approaches utilize a series of discrete characteristics of an item, in order to recommend...
