Recommendation System Simulations: A Discussion of Two Key Challenges

As recommendation systems become increasingly standard for online platforms, simulations provide an avenue for understanding the impacts of these systems on individuals and society. When constructing a recommendation system simulation, there are two key challenges: first, defining a model for users selecting or engaging with recommended items and second, defining a mechanism for users encountering items that are not recommended to the user directly by the platform, such as by a friend sharing specific content. This paper will delve into both of these challenges, reviewing simulation assumptions from existing research and proposing alternative assumptions. We also include a broader discussion of the limitations of simulations and outline of open questions in this area.

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

[2]  S. Dhami.  ECONOMICS OF INFORMATION , 2004 .

[3]  Barbara E. Engelhardt,et al.  How algorithmic confounding in recommendation systems increases homogeneity and decreases utility , 2017, RecSys.

[4]  Yehuda Koren,et al.  Factorization meets the neighborhood: a multifaceted collaborative filtering model , 2008, KDD.

[5]  Peter Boatwright,et al.  A Satisficing Choice Model , 2012, Mark. Sci..

[6]  B. Wernerfelt,et al.  An Evaluation Cost Model of Consideration Sets , 1990 .

[7]  Guy Aridor,et al.  Deconstructing the Filter Bubble: User Decision-Making and Recommender Systems , 2019, RecSys.

[8]  H. Simon,et al.  A Behavioral Model of Rational Choice , 1955 .

[9]  Arvind Narayanan,et al.  T-RECS: A Simulation Tool to Study the Societal Impact of Recommender Systems , 2021, ArXiv.

[10]  Jan Lorenz,et al.  The triple‐filter bubble: Using agent‐based modelling to test a meta‐theoretical framework for the emergence of filter bubbles and echo chambers , 2018, The British journal of social psychology.

[11]  Filippo Menczer,et al.  How algorithmic popularity bias hinders or promotes quality , 2017, Scientific Reports.

[12]  Duncan J. Watts,et al.  Estimating the Causal Impact of Recommendation Systems from Observational Data , 2015, EC.

[13]  John D. C. Little,et al.  A Logit Model of Brand Choice Calibrated on Scanner Data , 2011, Mark. Sci..

[14]  Steven M. Shugan.  The Cost Of Thinking , 1980 .

[15]  John Roberts,et al.  Development and Testing of a Model of Consideration Set Composition , 1991 .

[16]  Olfa Nasraoui,et al.  Debiasing the Human-Recommender System Feedback Loop in Collaborative Filtering , 2019, WWW.

[17]  John R. Hauser,et al.  Recommending Products When Consumers Learn Their Preference Weights , 2019, Mark. Sci..

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