CoFiRank - Maximum Margin Matrix Factorization for Collaborative Ranking

Markus Weimer (1) , Alexandros Karatzoglou (2) , Quoc Le (3) , Alex Smola (1)

1 Yahoo! Labs - Yahoo! Labs
2 TID - Telefonica Research
3 Computer Science Department [Stanford]

In this paper, we consider collaborative filtering as a ranking problem. We present a method which uses Maximum Margin Matrix Factorization and optimizes ranking instead of rating. We employ structured output prediction to optimize directly for ranking scores. Experimental results show that our method gives very good ranking scores and scales well on collaborative filtering tasks.

Mots clés: Collaborative Filtering; Ranking; Stuctured Estimation; NDCG

Domaines: Autres [stat.ML]

Markus Weimer, Alexandros Karatzoglou, Quoc Le, Alex Smola. CoFiRank - Maximum Margin Matrix Factorization for Collaborative Ranking. Advances in Neural Information Processing Systems, 21st Annual Conference on Neural Information Processing Systems 2007, Dec 2007, Vancouver, Canada. pp.222-230. ⟨hal-00482740⟩