Building Human Values into Recommender Systems: An Interdisciplinary Synthesis

Abstract

Recommender systems are the algorithms which select, filter, and personalize content across many of the worlds largest platforms and apps. As such, their positive and negative effects on individuals and on societies have been extensively theorized and studied. Our overarching question is how to ensure that recommender systems enact the values of the individuals and societies that they serve. Addressing this question in a principled fashion requires technical knowledge of recommender design and operation, and also critically depends on insights from diverse fields including social science, ethics, economics, psychology, policy and law. This paper is a multidisciplinary effort to synthesize theory and practice from different perspectives, with the goal of providing a shared language, articulating current design approaches, and identifying open problems. It is not a comprehensive survey of this large space, but a set of highlights identified by our diverse author cohort. We collect a set of values that seem most relevant to recommender systems operating across different domains, then examine them from the perspectives of current industry practice, measurement, product design, and policy approaches. Important open problems include multi-stakeholder processes for defining values and resolving trade-offs, better values-driven measurements, recommender controls that people use, non-behavioral algorithmic feedback, optimization for long-term outcomes, causal inference of recommender effects, academic-industry research collaborations, and interdisciplinary policy-making.

References in corpus (21)

• Explainable Recommendation: A Survey and New Perspectives
• Fairness of Exposure in Rankings
• A Survey on Conversational Recommender Systems
• A systematic review and taxonomy of explanations in decision support and recommender systems
• Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search
• Equity of Attention: Amortizing Individual Fairness in Rankings
• Beyond Personalization: Research Directions in Multistakeholder Recommendation
• Measurement and Fairness
• Measuring the Business Value of Recommender Systems
• Evaluating Stochastic Rankings with Expected Exposure
• Degenerate Feedback Loops in Recommender Systems
• Fairness in Information Access Systems
• Fairness in Ranking: A Survey
• The Structure of Toxic Conversations on Twitter
• PolicyKit: Building Governance in Online Communities
• User Fairness, Item Fairness, and Diversity for Rankings in Two-Sided Markets
• User Fairness in Recommender Systems
• Algorithmic Auditing and Social Justice: Lessons from the History of Audit Studies
• From Optimizing Engagement to Measuring Value
• IEEE 7010: A New Standard for Assessing the Well-being Implications of Artificial Intelligence
• Practitioners Versus Users: A Value-Sensitive Evaluation of Current Industrial Recommender System Design

Cited by in corpus (8)

• Recommender Systems for Good (RS4Good): Survey of Use Cases and a Call to Action for Research that Matters
• User Experience with LLM-powered Conversational Recommendation Systems: A Case of Music Recommendation
• Recommending With, Not For: Co-Designing Recommender Systems for Social Good
• Cura: Curation at Social Media Scale
• Explainable Multi-Stakeholder Job Recommender Systems
• Mapping Stakeholder Needs to Multi-Sided Fairness in Candidate Recommendation for Algorithmic Hiring
• External Evaluation of Discrimination Mitigation Efforts in Meta's Ad Delivery
• A Multistakeholder Approach to Value-Driven Co-Design of Recommender System Evaluation Metrics in Digital Archives