Moving beyond linearity and independence in top-N recommender systems

This paper suggests a number of research directions in which the recommender systems can improve their quality, by moving beyond the assumptions of linearity and independence that are traditionally made. These assumptions, while producing effective and meaningful results, can be suboptimal, as in lots of cases they do not represent the real datasets. In this paper, we discuss three different ways to address some of the previous constraints. More specifically, we focus on the development of methods capturing higher-order relations between the items, cross-feature interactions and intra-set dependencies which can potentially lead to a considerable enhancement of the recommendation accuracy.

[1]  Panagiotis Adamopoulos.  Beyond rating prediction accuracy: on new perspectives in recommender systems , 2013, RecSys.

[2]  Roberto Turrin,et al.  Performance of recommender algorithms on top-n recommendation tasks , 2010, RecSys '10.

[3]  Jennifer Golbeck,et al.  Curator: a game with a purpose for collection recommendation , 2010, CHI.

[4]  Yifan Hu,et al.  Collaborative Filtering for Implicit Feedback Datasets , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[5]  Lars Schmidt-Thieme,et al.  Online-updating regularized kernel matrix factorization models for large-scale recommender systems , 2008, RecSys '08.

[6]  Qiang Yang,et al.  One-Class Collaborative Filtering , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[7]  Gediminas Adomavicius,et al.  Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions , 2005, IEEE Transactions on Knowledge and Data Engineering.

[8]  Derek L. Hansen,et al.  Mixing it up: recommending collections of items , 2009, CHI.

[9]  Lars Schmidt-Thieme,et al.  BPR: Bayesian Personalized Ranking from Implicit Feedback , 2009, UAI.

[10]  George Karypis,et al.  Sparse linear methods with side information for top-n recommendations , 2012, WWW.

[11]  Deepak Agarwal,et al.  Regression-based latent factor models , 2009, KDD.

[12]  Geoffrey E. Hinton,et al.  Restricted Boltzmann machines for collaborative filtering , 2007, ICML '07.

[13]  Fabio Aiolli.  A Preliminary Study on a Recommender System for the Million Songs Dataset Challenge , 2013, IIR.

[14]  Alexander J. Smola,et al.  Like like alike: joint friendship and interest propagation in social networks , 2011, WWW.

[15]  George Karypis,et al.  Item-based top-N recommendation algorithms , 2004, TOIS.

[16]  Gediminas Adomavicius,et al.  Improving Aggregate Recommendation Diversity Using Ranking-Based Techniques , 2012, IEEE Transactions on Knowledge and Data Engineering.

[17]  Geoffrey J. Gordon,et al.  Relational learning via collective matrix factorization , 2008, KDD.

[18]  Evangelia Christakopoulou,et al.  HOSLIM: Higher-Order Sparse LInear Method for Top-N Recommender Systems , 2014, PAKDD.

[19]  George Karypis,et al.  SLIM: Sparse Linear Methods for Top-N Recommender Systems , 2011, 2011 IEEE 11th International Conference on Data Mining.