One-class collaborative filtering

Abstract

Many applications of collaborative filtering (CF), such as news item recommendation and bookmark rec- ommendation, are most naturally thought of as one- class collaborative filtering (OCCF) problems. In these problems, the training data usually consist simply of bi- nary data reflecting a user's action or inaction, such as page visitation in the case of news item recommenda- tion or webpage bookmarking in the bookmarking sce- nario. Usually this kind of data are extremely sparse (a small fraction are positive examples), therefore am- biguity arises in the interpretation of the non-positive examples. Negative examples and unlabeled positive ex- amples are mixed together and we are typically unable to distinguish them. For example, we cannot really at- tribute a user not bookmarking a page to a lack of inter- est or lack of awareness of the page. Previous research addressing this one-class problem only considered it as a classification task. In this paper, we consider the one- class problem under the CF setting. We propose two frameworks to tackle OCCF. One is based on weighted low rank approximation; the other is based on negative example sampling. The experimental results show that our approaches significantly outperform the baselines. © 2008 IEEE.