The Stereotyping Problem in Collaboratively Filtered Recommender Systems

Recommender systems play a crucial role in mediating our access to online information. We show that such algorithms induce a particular kind of stereotyping: if preferences for a set of items are anti-correlated in the general user population, then those items may not be recommended together to a user, regardless of that user’s preferences and rating history. First, we introduce a notion of joint accessibility, which measures the extent to which a set of items can jointly be accessed by users. We then study joint accessibility under the standard factorization-based collaborative filtering framework, and provide theoretical necessary and sufficient conditions when joint accessibility is violated. Moreover, we show that these conditions can easily be violated when the users are represented by a single feature vector. To improve joint accessibility, we further propose an alternative modelling fix, which is designed to capture the diverse multiple interests of each user using a multi-vector representation. We conduct extensive experiments on real and simulated datasets, demonstrating the stereotyping problem with standard single-vector matrix factorization models.

[1]  Inderjit S. Dhillon,et al.  Scalable Coordinate Descent Approaches to Parallel Matrix Factorization for Recommender Systems , 2012, 2012 IEEE 12th International Conference on Data Mining.

[2]  Blanca Vargas-Govea,et al.  Effects of relevant contextual features in the performance of a restaurant recommender system , 2011 .

[3]  MengChu Zhou,et al.  An Efficient Non-Negative Matrix-Factorization-Based Approach to Collaborative Filtering for Recommender Systems , 2014, IEEE Transactions on Industrial Informatics.

[4]  Karl Aberer,et al.  SoCo: a social network aided context-aware recommender system , 2013, WWW.

[5]  Michael I. Jordan,et al.  Do Offline Metrics Predict Online Performance in Recommender Systems? , 2020, ArXiv.

[6]  Matevz Kunaver,et al.  Diversity in recommender systems - A survey , 2017, Knowl. Based Syst..

[7]  Markus Zanker,et al.  Decision Biases in Recommender Systems , 2015 .

[8]  Kartik Hosanagar,et al.  Recommender systems and their impact on sales diversity , 2007, EC '07.

[9]  F. Maxwell Harper,et al.  The MovieLens Datasets: History and Context , 2016, TIIS.

[10]  Christophe Diot,et al.  Finding a needle in a haystack of reviews: cold start context-based hotel recommender system , 2012, RecSys.

[11]  Atsuhiro Takasu,et al.  Hybrid Recommender System Using Latent Features , 2009, 2009 International Conference on Advanced Information Networking and Applications Workshops.

[12]  Domonkos Tikk,et al.  Investigation of Various Matrix Factorization Methods for Large Recommender Systems , 2008, 2008 IEEE International Conference on Data Mining Workshops.

[13]  Farida Karimova,et al.  A Survey of e-Commerce Recommender Systems , 2016 .

[14]  Yi Tay,et al.  Deep Learning based Recommender System: A Survey and New Perspectives , 2018 .

[15]  Marcelo Mendoza,et al.  Evaluating content novelty in recommender systems , 2019, Journal of Intelligent Information Systems.

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

[17]  Tor Lattimore,et al.  Degenerate Feedback Loops in Recommender Systems , 2019, AIES.

[18]  R. Ravi,et al.  Post Processing Recommender Systems for Diversity , 2017, KDD.

[19]  Rui Zhang,et al.  Security and Privacy on Blockchain , 2019, ACM Comput. Surv..

[20]  Osmar R. Zaïane,et al.  Building a Recommender Agent for e-Learning Systems , 2002, ICCE.

[21]  Licia Capra,et al.  Temporal diversity in recommender systems , 2010, SIGIR.

[22]  Jonathan L. Herlocker,et al.  Evaluating collaborative filtering recommender systems , 2004, TOIS.

[23]  Mehrbakhsh Nilashi,et al.  Collaborative filtering recommender systems , 2013 .

[24]  Shujian Huang,et al.  Deep Matrix Factorization Models for Recommender Systems , 2017, IJCAI.

[25]  Chang Zhou,et al.  Controllable Multi-Interest Framework for Recommendation , 2020, KDD.

[26]  Hao Wu,et al.  Dual-regularized matrix factorization with deep neural networks for recommender systems , 2018, Knowl. Based Syst..

[27]  Damian Trilling,et al.  Do not blame it on the algorithm: an empirical assessment of multiple recommender systems and their impact on content diversity , 2018 .

[28]  F. Frances Yao,et al.  Computational Geometry , 1991, Handbook of Theoretical Computer Science, Volume A: Algorithms and Complexity.

[29]  Fernando Ortega,et al.  A non negative matrix factorization for collaborative filtering recommender systems based on a Bayesian probabilistic model , 2016, Knowl. Based Syst..

[30]  Loren G. Terveen,et al.  Exploring the filter bubble: the effect of using recommender systems on content diversity , 2014, WWW.

[31]  Yongfeng Zhang,et al.  Understanding Echo Chambers in E-commerce Recommender Systems , 2020, SIGIR.

[32]  Alejandro Bellogín,et al.  Statistical biases in Information Retrieval metrics for recommender systems , 2017, Information Retrieval Journal.

[33]  Paolo Tomeo,et al.  Adaptive multi-attribute diversity for recommender systems , 2017, Inf. Sci..

[34]  Lucia D'Acunto,et al.  Exposure diversity as a design principle for recommender systems , 2018 .

[35]  John Riedl,et al.  Recommender systems in e-commerce , 1999, EC '99.

[36]  Robin Burke,et al.  Context-aware music recommendation based on latenttopic sequential patterns , 2012, RecSys.

[37]  Vittorio Loreto,et al.  Towards novelty-driven recommender systems , 2019, Comptes Rendus Physique.

[38]  Benjamin Recht,et al.  Recommendations and user agency: the reachability of collaboratively-filtered information , 2020, FAT*.

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