Incremental Matrix Co-factorization for Recommender Systems with Implicit Feedback

Recommender systems try to predict which items a user will prefer. Traditional models for recommendation only take into account the user-item interaction, usually expressed by explicit ratings. However, in these days, web services continuously generate auxiliary data from users and items that can be incorporated into the recommendation model to improve recommendations. In this work, we propose an incremental Matrix Co-factorization model with implicit user feedback, considering a real-world data-stream scenario. This model can be seen as an extension of the conventional Matrix Factorization that includes additional dimensions to be decomposed in the common latent factor space. We test our proposal against a baseline algorithm that relies exclusively on interaction data, using prequential evaluation. Our experimental results show a significant improvement in the accuracy of recommendations, after incorporating an additional dimension in three music domain datasets.

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

[2]  Marcos Aurélio Domingues,et al.  Dimensions as Virtual Items: Improving the predictive ability of top-N recommender systems , 2013, Inf. Process. Manag..

[3]  João Gama,et al.  Evaluation of recommender systems in streaming environments , 2015, ArXiv.

[4]  Yehuda Koren,et al.  Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights , 2007, Seventh IEEE International Conference on Data Mining (ICDM 2007).

[5]  Yanxiang Huang,et al.  TencentRec: Real-time Stream Recommendation in Practice , 2015, SIGMOD Conference.

[6]  Zi Huang,et al.  Joint Modeling of User Check-in Behaviors for Real-time Point-of-Interest Recommendation , 2016, ACM Trans. Inf. Syst..

[7]  Balázs Hidasi,et al.  Fast ALS-based tensor factorization for context-aware recommendation from implicit feedback , 2012, ECML/PKDD.

[8]  John Riedl,et al.  Incremental SVD-Based Algorithms for Highly Scaleable Recommender Systems , 2002 .

[9]  James Caverlee,et al.  Multi-Aspect Streaming Tensor Completion , 2017, KDD.

[10]  João Gama,et al.  Fast Incremental Matrix Factorization for Recommendation with Positive-Only Feedback , 2014, UMAP.

[11]  Eva Zangerle,et al.  #nowplaying Music Dataset: Extracting Listening Behavior from Twitter , 2014, WISMM '14.

[12]  João Gama,et al.  Issues in evaluation of stream learning algorithms , 2009, KDD.

[13]  Domonkos Tikk,et al.  Scalable Collaborative Filtering Approaches for Large Recommender Systems , 2009, J. Mach. Learn. Res..

[14]  Xing Xie,et al.  Learning travel recommendations from user-generated GPS traces , 2011, TIST.

[15]  João Gama,et al.  Collaborative filtering with recency-based negative feedback , 2015, SAC.

[16]  James Bailey,et al.  Accelerating Online CP Decompositions for Higher Order Tensors , 2016, KDD.

[17]  Luo Si,et al.  Matrix co-factorization for recommendation with rich side information and implicit feedback , 2011, HetRec '11.

[18]  Fangfang Li,et al.  Coupled Item-Based Matrix Factorization , 2014, WISE.

[19]  Òscar Celma,et al.  Music Recommendation and Discovery - The Long Tail, Long Fail, and Long Play in the Digital Music Space , 2010 .

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

[21]  Nuria Oliver,et al.  Multiverse recommendation: n-dimensional tensor factorization for context-aware collaborative filtering , 2010, RecSys '10.

[22]  Hiroyuki Kasai,et al.  Online low-rank tensor subspace tracking from incomplete data by CP decomposition using recursive least squares , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[23]  Brian D. Davison,et al.  Co-factorization machines: modeling user interests and predicting individual decisions in Twitter , 2013, WSDM.

[24]  Peter J. Haas,et al.  Large-scale matrix factorization with distributed stochastic gradient descent , 2011, KDD.

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

[26]  András A. Benczúr,et al.  Exploiting temporal influence in online recommendation , 2014, RecSys '14.

[27]  Badrish Chandramouli,et al.  StreamRec: a real-time recommender system , 2011, SIGMOD '11.

[28]  Domonkos Tikk,et al.  Fast als-based matrix factorization for explicit and implicit feedback datasets , 2010, RecSys '10.

[29]  Yehuda Koren,et al.  Matrix Factorization Techniques for Recommender Systems , 2009, Computer.