TY - GEN
T1 - Local collaborative ranking
AU - Lee, Joonseok
AU - Bengio, Samy
AU - Kim, Seungyeon
AU - Lebanon, Guy
AU - Singer, Yoram
PY - 2014/4/7
Y1 - 2014/4/7
N2 - Personalized recommendation systems are used in a wide variety of applications such as electronic commerce, social networks, web search, and more. Collaborative filtering approaches to recommendation systems typically assume that the rating matrix (e.g., movie ratings by viewers) is low- rank. In this paper, we examine an alternative approach in which the rating matrix is locally low-rank. Concretely, we assume that the rating matrix is low-rank within certain neighborhoods of the metric space defined by (user, item) pairs. We combine a recent approach for local low-rank approximation based on the Frobenius norm with a general empirical risk minimization for ranking losses. Our experiments indicate that the combination of a mixture of local low-rank matrices each of which was trained to minimize a ranking loss outperforms many of the currently used state- of-the-art recommendation systems. Moreover, our method is easy to parallelize, making it a viable approach for large scale real-world rank-based recommendation systems. Copyright is held by the International World Wide Web Conference Committee (IW3C2).
AB - Personalized recommendation systems are used in a wide variety of applications such as electronic commerce, social networks, web search, and more. Collaborative filtering approaches to recommendation systems typically assume that the rating matrix (e.g., movie ratings by viewers) is low- rank. In this paper, we examine an alternative approach in which the rating matrix is locally low-rank. Concretely, we assume that the rating matrix is low-rank within certain neighborhoods of the metric space defined by (user, item) pairs. We combine a recent approach for local low-rank approximation based on the Frobenius norm with a general empirical risk minimization for ranking losses. Our experiments indicate that the combination of a mixture of local low-rank matrices each of which was trained to minimize a ranking loss outperforms many of the currently used state- of-the-art recommendation systems. Moreover, our method is easy to parallelize, making it a viable approach for large scale real-world rank-based recommendation systems. Copyright is held by the International World Wide Web Conference Committee (IW3C2).
KW - Collaborative filtering
KW - Ranking
KW - Recommender systems
UR - http://www.scopus.com/inward/record.url?scp=84908666141&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84908666141&partnerID=8YFLogxK
U2 - 10.1145/2566486.2567970
DO - 10.1145/2566486.2567970
M3 - Conference contribution
AN - SCOPUS:84908666141
T3 - WWW 2014 - Proceedings of the 23rd International Conference on World Wide Web
SP - 85
EP - 95
BT - WWW 2014 - Proceedings of the 23rd International Conference on World Wide Web
PB - Association for Computing Machinery
T2 - 23rd International Conference on World Wide Web, WWW 2014
Y2 - 7 April 2014 through 11 April 2014
ER -