Building Human Values into Recommender Systems: An Interdisciplinary Synthesis

@article{Stray2022BuildingHV,
  title={Building Human Values into Recommender Systems: An Interdisciplinary Synthesis},
  author={Jonathan Stray and Alon Y. Halevy and Parisa Assar and Dylan Hadfield-Menell and Craig Boutilier and Amar Ashar and Lex Beattie and Michael D. Ekstrand and Claire Leibowicz and Connie Moon Sehat and Sara Johansen and Lianne Kerlin and David Vickrey and Spandana Singh and Sanne Vrijenhoek and Amy X. Zhang and McKane Andrus and Natali Helberger and Polina Proutskova and Tanushree Mitra and Nina Vasan},
  journal={ArXiv},
  year={2022},
  volume={abs/2207.10192}
}
Recommender systems are the algorithms which select, filter, and personalize content across many of the world’s 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… 
1 Citations

Figures from this paper

RecSys Fairness Metrics: Many to Use But Which One To Choose?

TLDR
This position paper focuses on a widening gap between the research community and practitioners concerning the availability of metrics versus the ability to put them into practice, and develops methods to help ML practitioners in their decision-making processes when picking fairness metrics for recommendation and ranking systems.

References

SHOWING 1-10 OF 423 REFERENCES

Assessing the Value of Transparency in Recommender Systems: An End-User Perspective

TLDR
This work investigated how end-users of recommender systems value different categories of information to help in determining what to do with computer-generated recommendations in contexts involving high risk to themselves or others.

Multistakeholder recommendation: Survey and research directions

TLDR
The multistakeholder perspective on recommendation is outlined, highlighting example research areas and discussing important issues, open questions, and prospective research directions.

Recommenders with a Mission: Assessing Diversity in News Recommendations

TLDR
This paper aims to bridge the gap between normative notions of diversity, rooted in democratic theory, and quantitative metrics necessary for evaluating the recommender system, and proposes a set of metrics grounded in social science interpretations of diversity.

Do not blame it on the algorithm: an empirical assessment of multiple recommender systems and their impact on content diversity

TLDR
It is found that all of the recommendation logics under study proved to lead to a rather diverse set of recommendations that are on par with human editors and that basing recommendations on user histories can substantially increase topic diversity within a recommendation set.

Measuring the Business Value of Recommender Systems

TLDR
A review of existing publications on field tests of recommender systems and which business-related performance measures were used in such real-world deployments indicates that various open questions remain regarding the realistic quantification of the business effects of recommenders and the performance assessment of recommendation algorithms in academia.

Designing for the better by taking users into account: a qualitative evaluation of user control mechanisms in (news) recommender systems

TLDR
This paper focuses and elaborate on the concept of user control because that mitigates multiple problems at once and contributes to a richer understanding of why and how to design for user control in recommender systems.

A user-centric evaluation framework for recommender systems

TLDR
A unifying evaluation framework, called ResQue (Recommender systems' Quality of user experience), which aimed at measuring the qualities of the recommended items, the system's usability, usefulness, interface and interaction qualities, users' satisfaction with the systems, and the influence of these qualities on users' behavioral intentions.

Exposure diversity as a design principle for recommender systems

TLDR
Suggestions for design principles are made and the potential and possible limits of ‘diversity sensitive design’ are explored, to break potential ‘filter bubbles’ rather than create them.

Recommender Systems and Misinformation: The Problem or the Solution?

TLDR
Some of the key challenges behind assessing and measuring the effect of existing recommendation algorithms on the recommendation of misinforming articles and how such algorithms could be adapted, modified, and evaluated to counter this effect based on existing social science and psychology research are described.

A systematic review and taxonomy of explanations in decision support and recommender systems

TLDR
This work systematically review the literature on explanations in advice-giving systems, which includes recommender systems, and derives a novel comprehensive taxonomy of aspects to be considered when designing explanation facilities for current and future decision support systems.
...