One of our areas of interest is bias and discrimination in recommender systems. As machine learning, data mining, and other artificial intelligence techniques become increasingly pervasive in our daily lives, the research community has started to turn our attention to the question of whether they are fair. There is a variety of work happening on algorithmic fairness to try to assess whether particular systems are unfair or discriminatory, and how to mitigate this situation.

In this project, we are investigating several questions of fairness and bias in recommender systems:

  • What does it mean for a recommender to be fair, unfair, or biased?
  • What potentially discriminatory biases are present in the recommender’s input data, algorithmic structure, or output?
  • How do these biases change over time through the recommender-user feedback loop?

This is a part of our overall, ongoing goal to help make recommenders (and other AI systems) better for the people they affect.

Publications

Christine Pinney, Amifa Raj, Alex Hanna, and Michael D. Ekstrand. 2023. “Much Ado About Gender: Current Practices and Future Recommendations for Appropriate Gender-Aware Information Access”. In ACM SIGIR Conference on Human Information Interaction and Retrieval (CHIIR ‘23). DOI:10.1145/3576840.3578316.

Jonathan Stray, Alon Halevy, Parisa Assar, Dylan Hadfield-Menell, Craig Boutilier, Amar Ashar, Lex Beattie, Michael Ekstrand, Claire Leibowicz, Connie Moon Sehat, Sara Johansen, Lianne Kerlin, David Vickrey, Spandana Singh, Sanne Vrijenhoek, Amy Zhang, Mckane Andrus, Natali Helberger, Polina Proutskova, Tanushree Mitra, and Nina Vasan. 2022. Building Human Values into Recommender Systems: An Interdisciplinary Synthesis. Preprint manuscript arXiv:2207.10192 [cs.IR].

Amifa Raj, Michael D. Ekstrand. 2022. “Fire Dragon and Unicorn Princess; Gender Stereotypes and Children’s Products in Search Engine Responses”. In Proceedings of SIGIR ecom’22: ACM SIGIR Workshop on eCommerce.

Amifa Raj, Michael D. Ekstrand. 2022. “Measuring Fairness in Ranked Results: An Analytical and Empirical Comparison”. In SIGIR ‘22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval(July 2022), 726-736. DOI:10.1145/3477495.3532018.

Michael D. Ekstrand, Anubrata Das, Robin Burke, and Fernando Diaz. 2022. “Fairness and Discrimination in Information Access Systems”. Foundations and Trends® in Information Retrieval 16(1–2), 1–177. DOI:10.1561/1500000079.

Nasim Sonboli, Robin Burke, Michael Ekstrand, and Rishabh Mehrotra. 2022. The Multisided Complexity of Fairness in Recommender Systems. AI Magazine 43 (June 2022), 164–176. DOI:10.1002/aaai.12054.

Michael D. Ekstrand, Anubrata Das, Robin Burke, and Fernando Diaz. 2021. “Fairness in Recommender Systems”. In Recommender System Handbook, 3rd edition. Francesco Ricci, Lior Roach, and Bracha Shapira, eds. Springer-Verlag. ISBN 978-1-0716-2196-7. DOI:10.1007/978-1-0716-2197-4_18.

Amifa Raj, Ashlee Milton, Michael D. Ekstrand. 2021. “Pink for Princesses, Blue for Superheroes: The Need to Examine Gender Stereotypes in Kid’s Products in Search and Recommendations”. In Proceedings of the 5th International and Interdisciplinary Perspectives on Children and Recommender and Information Retrieval Systems: Search and Recommendation Technology through the Lens of a Teacher (KidRec 2021), co-located with ACM IDC 2021.

Ömer Kırnap, Fernando Diaz, Asia J. Biega, Michael D. Ekstrand, Ben Carterette, and Emine Yılmaz. 2021. Estimation of Fair Ranking Metrics with Incomplete Judgments. In Proceedings of The Web Conference 2021 (TheWebConf 2021). DOI:10.1145/3442381.3450080.

Michael D. Ekstrand and Daniel Kluver. 2021. Exploring Author Gender in Book Rating and Recommendation. User Modeling and User-Adapted Interaction 31(3) (2021). DOI:10.1145/3442381.3450080.

Amifa Raj, Connor Wood, Ananda Montoly and Michael D. Ekstrand. 2020. “Comparing Fair Ranking Metrics”. To appear in Proceedings of the 3rd FAccTRec Workshop on Responsible Recommendation at 14th ACM Conference on Recommender Systems (RecSys 20).

Asia J. Biega, Fernando Diaz, Michael D. Ekstrand, and Sebastian Kohlmeier. 2020. Overview of the TREC 2019 Fair Ranking Track. In The Twenty-Eighth Text REtrieval Conference (TREC 2019) Proceedings. arXiv:2003.11650.

Michael D. Ekstrand, Mucun Tian, Mohammed R. Imran Kazi, Hoda Mehrpouyan, and Daniel Kluver. 2018. “Exploring Author Gender in Book Rating and Recommendation”. In Proceedings of the 12th ACM Conference on Recommender Systems (RecSys 2018).

Michael D. Ekstrand, Mucun Tian, Ion Madrazo Azpiazu, Jennifer D. Ekstrand, Oghenemaro Anuyah, David McNeill, and Maria Soledad Pera. 2018. “All The Cool Kids, How Do They Fit In?: Popularity and Demographic Biases in Recommender Evaluation and Effectiveness”. In Proceedings of the 1st Conference on Fairness, Accountability and Transparency (FAT* 2018).

Michael D. Ekstrand and Maria Soledad Pera. 2017. “The Demographics of Cool: Popularity and Recommender Performance for Different Groups of Users”. In Proceedings of the 11th ACM Conference on Recommender Systems (RecSys 2017 Poster Proceedings).

Sushma Channamsetty and Michael D. Ekstrand. 2017. “Recommender Response to Diversity and Popularity Bias in User Profiles”. In Proceedings of the 30th International Florida Artificial Intelligence Research Society Conference (FLAIRS 2017).