Popularity-Opportunity Bias in Collaborative Filtering

This paper connects equal opportunity to popularity bias in implicit recommenders to introduce the problem of popularity-opportunity bias. That is, conditioned on user preferences that a user likes both items, the more popular item is more likely to be recommended (or ranked higher) to the user than the less popular one. This type of bias is harmful, exerting negative effects on the engagement of both users and item providers. Thus, we conduct a three-part study: (i) By a comprehensive empirical study, we identify the existence of the popularity-opportunity bias in fundamental matrix factorization models on four datasets; (ii) coupled with this empirical study, our theoretical study shows that matrix factorization models inherently produce the bias; and (iii) we demonstrate the potential of alleviating this bias by both in-processing and post-processing algorithms. Extensive experiments on four datasets show the effective debiasing performance of these proposed methods compared with baselines designed for conventional popularity bias.

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

[2]  James Caverlee,et al.  Next-item Recommendation with Sequential Hypergraphs , 2020, SIGIR.

[3]  Zhe Zhao,et al.  Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations , 2017, ArXiv.

[4]  Nathan Srebro,et al.  Equality of Opportunity in Supervised Learning , 2016, NIPS.

[5]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[6]  Priscilla S. Markwood,et al.  The Long Tail: Why the Future of Business is Selling Less of More , 2006 .

[7]  Robin Burke,et al.  The Unfairness of Popularity Bias in Recommendation , 2019, RMSE@RecSys.

[8]  Harald Steck,et al.  Item popularity and recommendation accuracy , 2011, RecSys '11.

[9]  Li Chen,et al.  How Serendipity Improves User Satisfaction with Recommendations? A Large-Scale User Evaluation , 2019, WWW.

[10]  Dietmar Jannach,et al.  What recommenders recommend: an analysis of recommendation biases and possible countermeasures , 2015, User Modeling and User-Adapted Interaction.

[11]  Bamshad Mobasher,et al.  Controlling Popularity Bias in Learning-to-Rank Recommendation , 2017, RecSys.

[12]  Bamshad Mobasher,et al.  Managing Popularity Bias in Recommender Systems with Personalized Re-ranking , 2019, FLAIRS.

[13]  Ed H. Chi,et al.  Toward a better trade-off between performance and fairness with kernel-based distribution matching , 2019, ArXiv.

[14]  Pedro Cano,et al.  From hits to niches?: or how popular artists can bias music recommendation and discovery , 2008, NETFLIX '08.

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

[16]  James Caverlee,et al.  Measuring and Mitigating Item Under-Recommendation Bias in Personalized Ranking Systems , 2020, SIGIR.

[17]  Anton van den Hengel,et al.  Image-Based Recommendations on Styles and Substitutes , 2015, SIGIR.

[18]  Yuhong Guo,et al.  Improving Top-N Recommendation with Heterogeneous Loss , 2016, IJCAI.

[19]  James Caverlee,et al.  Fairness-Aware Tensor-Based Recommendation , 2018, CIKM.

[20]  Huan Liu,et al.  mTrust: discerning multi-faceted trust in a connected world , 2012, WSDM '12.

[21]  Yu Jeffrey Hu,et al.  From Niches to Riches: Anatomy of the Long Tail , 2006 .

[22]  James Caverlee,et al.  Recommending Music Curators: A Neural Style-Aware Approach , 2020, ECIR.

[23]  Julian J. McAuley,et al.  VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback , 2015, AAAI.

[24]  Shotaro Akaho,et al.  Considerations on Recommendation Independence for a Find-Good-Items Task , 2017 .

[25]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[26]  Jun Sakuma,et al.  Recommendation Independence , 2018, FAT.

[27]  David M. Blei,et al.  Factorization Meets the Item Embedding: Regularizing Matrix Factorization with Item Co-occurrence , 2016, RecSys.

[28]  Sahin Cem Geyik,et al.  Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search , 2019, KDD.

[29]  Jun Sakuma,et al.  Correcting Popularity Bias by Enhancing Recommendation Neutrality , 2014, RecSys Posters.

[30]  Jun Sakuma,et al.  Efficiency Improvement of Neutrality-Enhanced Recommendation , 2013, Decisions@RecSys.

[31]  Harald Steck.  Collaborative Filtering via High-Dimensional Regression , 2019, ArXiv.

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

[33]  Robin Burke,et al.  Reducing Popularity Bias in Recommendation Over Time , 2019, ArXiv.

[34]  Tat-Seng Chua,et al.  Neural Collaborative Filtering , 2017, WWW.

[35]  Weiwen Liu,et al.  Personalizing Fairness-aware Re-ranking , 2018, ArXiv.

[36]  Blake Lemoine,et al.  Mitigating Unwanted Biases with Adversarial Learning , 2018, AIES.

[37]  Alexander Tuzhilin,et al.  The long tail of recommender systems and how to leverage it , 2008, RecSys '08.

[38]  Bert Huang,et al.  Beyond Parity: Fairness Objectives for Collaborative Filtering , 2017, NIPS.

[39]  Ed H. Chi,et al.  Fairness in Recommendation Ranking through Pairwise Comparisons , 2019, KDD.

[40]  Krishna P. Gummadi,et al.  Fighting Fire with Fire: Using Antidote Data to Improve Polarization and Fairness of Recommender Systems , 2018, WSDM.

[41]  Guy Shani,et al.  Evaluating Recommendation Systems , 2011, Recommender Systems Handbook.