Impacts of Mainstream-Driven Algorithms on Recommendations for Children Across Domains: A Reproducibility Study

2025-07-09

AI Summary

This study identifies systematic biases in mainstream recommendation algorithms toward child users across movie, music, and book domains. Prior work, hindered by scarce child-user data, fails to discern preference disparities between children and adults or quantify popularity bias effects. To address this, we first replicate and extend behavioral analysis of child recommendations on large-scale, multi-domain datasets, proposing a reproducible cross-domain evaluation framework that jointly incorporates standard recommendation metrics and a novel popularity-bias measurement. Experiments reveal that children’s interaction patterns differ significantly from those of adults, and algorithmic performance is highly domain-dependent; moreover, popularity bias not only degrades recommendation diversity but also exacerbates fairness disparities across age groups. Our core contributions are: (1) establishing the first cross-domain benchmarking paradigm for child-oriented recommendation research, and (2) empirically uncovering the differential impact mechanism of popularity bias on child users.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Application Domains: Humanities & Computational Social Science

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Children are often exposed to items curated by recommendation algorithms. Yet, research seldom considers children as a user group, and when it does, it is anchored on datasets where children are underrepresented, risking overlooking their interests, favoring those of the majority, i.e., mainstream users. Recently, Ungruh et al. demonstrated that children's consumption patterns and preferences differ from those of mainstream users, resulting in inconsistent recommendation algorithm performance and behavior for this user group. These findings, however, are based on two datasets with a limited child user sample. We reproduce and replicate this study on a wider range of datasets in the movie, music, and book domains, uncovering interaction patterns and aspects of child-recommender interactions consistent across domains, as well as those specific to some user samples in the data. We also extend insights from the original study with popularity bias metrics, given the interpretation of results from the original study. With this reproduction and extension, we uncover consumption patterns and differences between age groups stemming from intrinsic differences between children and others, and those unique to specific datasets or domains.

Problem

• Examining how mainstream-driven algorithms affect children's recommendations across domains
• Investigating reproducibility of child-specific recommendation patterns in diverse datasets
• Analyzing age-based consumption differences and popularity bias in recommendations

Innovation

• Reproduce study on diverse datasets
• Extend insights with popularity bias metrics
• Uncover domain-specific child-recommender interactions