Fair Recommender Systems

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.

Blog Posts and Other Coverage

Author Gender in Book Recommendations (30 Jul 2018)
Presenting at FAT* (1 Feb 2018)

Funding

2018–2023: NSF award 1751278, $482,081: CAREER: User-Based Simulation Methods for Quantifying Sources of Error and Bias in Recommender Systems

Workshops and Meetings

I have been involved with several workshops, sessions, etc. related to fair recommendation.

Fair Recommendation & Retrieval tutorial
Fairness track at TREC 2019
FATREC2 at RecSys 2018
Recommenders Session at FAT* 2018
FATREC at RecSys 2017, the Workshop on Responsible Recommendation (proceedings)

Publications

Michael D. Ekstrand and Daniel Kluver. 2021. Exploring Author Gender in Book Rating and Recommendation. User Modeling and User-Adapted Interaction (February 2021). DOI 10.1007/s11257-020-09284-2. NSF PAR 10218853. Cited 1 times.

Ömer Kırnap, Fernando Diaz, Asia J. Biega, Michael D. Ekstrand, Ben Carterette, and Emine Yılmaz. 2021. Estimation of Fair Ranking Metrics with Incomplete Judgments. To appear in Proceedings of The Web Conference 2021 (TheWebConf 2021). ACM. DOI 10.1145/3442381.3450080. Acceptance rate: 21%.

Fernando Diaz, Bhaskar Mitra, Michael D. Ekstrand, Asia J. Biega, and Ben Carterette. 2020. Evaluating Stochastic Rankings with Expected Exposure. In Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM '20). ACM, pp. 275–284. DOI 10.1145/3340531.3411962. arXiv:2004.13157 [cs.IR]. NSF PAR 10199451. Acceptance rate: 20%.Nominated for Best Long Paper. Cited 6 times.

Amifa Raj, Connor Wood, Ananda Montoly, and Michael D. Ekstrand. 2020. Comparing Fair Ranking Metrics. Presented at the 3rd FAccTrec Workshop on Responsible Recommendation. arXiv:2009.01311 [cs.IR]. Cited 1 times.

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 (TREC 2019). arXiv:2003.11650.

Alexandra Olteanu, Jean Garcia-Gathright, Maarten de Rijke, Michael D. Ekstrand, Adam Roegiest, Aldo Lipani, Alex Beutel, Ana Lucic, Ana-Andreea Stoica, Anubrata Das, Asia Biega, Bart Voorn, Claudia Hauff, Damiano Spina, David Lewis, Douglas W Oard, Emine Yilmaz, Faegheh Hasibi, Gabriella Kazai, Graham McDonald, Hinda Haned, Iadh Ounis, Ilse van der Linden, Joris Baan, Kamuela N Lau, Krisztian Balog, Mahmoud Sayed, Maria Panteli, Mark Sanderson, Matthew Lease, Preethi Lahoti, and Toshihiro Kamishima. 2019. FACTS-IR: Fairness, Accountability, Confidentiality, Transparency, and Safety in Information Retrieval. SIGIR Forum 53(2) (December 2019), 20–43. DOI 10.1145/3458553.3458556. Cited 2 times.

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 '18). ACM, pp. 242–250. DOI 10.1145/3240323.3240373. arXiv:1808.07586v1 [cs.IR]. Acceptance rate: 17.5%. Cited 43 times.

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). PMLR, Proceedings of Machine Learning Research 81:172–186. Acceptance rate: 24%. Cited 33 times.

Michael D. Ekstrand and Maria Soledad Pera. 2017. The Demographics of Cool: Popularity and Recommender Performance for Different Groups of Users. In RecSys 2017 Poster Proceedings. CEUR, Workshop Proceedings 1905. Cited 3 times.

Workshop Summaries

Alexandra Olteanu, Jean Garcia-Gathright, Maarten de Rijke, and Michael D. Ekstrand. 2019. Workshop on Fairness, Accountability, Confidentiality, Transparency, and Safety in Information Retrieval (FACTS-IR). In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '19). ACM. DOI 10.1145/3331184.3331644. Cited 2 times.

Unresolved citation fairumap2

Toshihiro Kamishima, Pierre-Nicolas Schwab, and Michael D. Ekstrand. 2018. 2nd FATREC Workshop: Responsible Recommendation. In Proceedings of the 12th ACM Conference on Recommender Systems (RecSys '18). ACM. DOI 10.1145/3240323.3240335. Cited 3 times.

Bamshad Mobasher, Robin Burke, Michael D. Ekstrand, and Bettina Berendt. 2018. UMAP 2018 Fairness in User Modeling, Adaptation and Personalization (FairUMAP 2018) Chairs' Welcome & Organization. In Adjunct Publication of the 26th Conference on User Modeling, Adaptation, and Personalization (UMAP '18). ACM. DOI 10.1145/3213586.3226200.

Michael D. Ekstrand and Amit Sharma. 2017. The FATREC Workshop on Responsible Recommendation. In Proceedings of the 11th ACM Conference on Recommender Systems (RecSys '17). ACM. DOI 10.1145/3109859.3109960. Cited 6 times.