It's Not You, It's Me: The Impact of Choice Models and Ranking Strategies on Gender Imbalance in Music Recommendation

TL;DR

This work simulates user interaction and re-training to investigate the effects of ranking strategies and user choice models on gender fairness metrics and finds re-ranking strategies have a greater effect than user choice models on recommendation fairness over time.

Abstract

As recommender systems are prone to various biases, mitigation approaches are needed to ensure that recommendations are fair to various stakeholders. One particular concern in music recommendation is artist gender fairness. Recent work has shown that the gender imbalance in the sector translates to the output of music recommender systems, creating a feedback loop that can reinforce gender biases over time. In this work, we examine that feedback loop to study whether algorithmic strategies or user behavior are a greater contributor to ongoing improvement (or loss) in fairness as models are repeatedly re-trained on new user feedback data. We simulate user interaction and re-training to investigate the effects of ranking strategies and user choice models on gender fairness metrics. We find re-ranking strategies have a greater effect than user choice models on recommendation fairness over time.

Figures (3)

• Figure 1: Average position of the first female artist ($PFA$) over iterations. Position $0$ is the highest rank.
• Figure 2: $AWRF$ over iterations. $0$ is fair; negative values mean that unprotected-group (male) artists are overexposed.
• Figure 3: Gini coefficients ($Gini@10$) over iterations, both overall and for male and female artists. $1$ is maximum inequality.