Description

Collaborative Filtering is an important and popular technology for recommender systems. These methods are classified into user-based CF and item-based CF. The basic idea of user based CF approach is to find out set of users who have similar favor patterns to a given user (i.e., “neighbors” of the user) and recommend to the user those items that other users in the same set like, while the item-based CF approach aims to provide a user with the recommendation on an item based on the other items with high correlations (i.e., “neighbors” of the item). In all collaborative filtering methods, it is a significant step to find users’ (or items’) neighbors, that is, a set of similar users (or items). Currently, almost all CF methods measure users’ similarity (or items’ similarity) based on corated items of users (or common users of items). In this we propose a typicality-based collaborative filtering approach named TyCo, in which the “neighbors” of users are found based on user typicality in user groups instead of co-rated items of users. It investigates the collaborative filtering recommendation from a new perspective and presents a novel typicality-based collaborative filtering recommendation method named TyCo. In TyCo, a user is represented by a user typicality vector that can indicate the user’s preference on each kind of items. A distinct feature of TyCo is that it selects “neighbors” of users by measuring users’ similarity based on their typicality degrees instead of corated items by users. Such a feature can overcome several limitations of traditional collaborative filtering methods. It is the first work that applies typicality for collaborative filtering. We conduct experiments to evaluate TyCo and demonstrate the advantages of TyCo. In TyCo, there are some preprocessing procedures, such as constructing user prototype by clustering and measuring user typicality in user groups. The cost of these preprocessing procedures depends on the particular clustering method used.