Measuring Recommender System Effects with Simulated Users

TL;DR

This paper introduces a simulation framework to measure how recommender systems influence user behavior and biases over time, enabling analysis of effects beyond single interactions and across diverse user types.

Contribution

It presents a novel simulation framework and evaluation metrics to isolate recommender effects from user preferences and study long-term impacts and biases.

Findings

Popularity bias persists over time in both traditional and production systems.

Simulation reveals how biases influence user experience and system performance.

Framework helps identify effects of recommender choices on atypical user behaviors.

Abstract

Imagine a food recommender system -- how would we check if it is \emph{causing} and fostering unhealthy eating habits or merely reflecting users' interests? How much of a user's experience over time with a recommender is caused by the recommender system's choices and biases, and how much is based on the user's preferences and biases? Popularity bias and filter bubbles are two of the most well-studied recommender system biases, but most of the prior research has focused on understanding the system behavior in a single recommendation step. How do these biases interplay with user behavior, and what types of user experiences are created from repeated interactions? In this work, we offer a simulation framework for measuring the impact of a recommender system under different types of user behavior. Using this simulation framework, we can (a) isolate the effect of the recommender system from the user preferences, and (b) examine how the system performs not just on average for an "average user" but also the extreme experiences under atypical user behavior. As part of the simulation framework, we propose a set of evaluation metrics over the simulations to understand the recommender system's behavior. Finally, we present two empirical case studies -- one on traditional collaborative filtering in MovieLens and one on a large-scale production recommender system -- to understand how popularity bias manifests over time.