FairBayRank: A Fair Personalized Bayesian Ranker

Armielle Noulapeu Ngaffo; Julien Albert; Benoît Frénay; Gilles Perrouin

FairBayRank: A Fair Personalized Bayesian Ranker

Armielle Noulapeu Ngaffo, Julien Albert, Benoît Frénay, Gilles Perrouin

Research output: Contribution in Book/Catalog/Report/Conference proceeding › Conference contribution

21 Downloads (Pure)

Abstract

Recommender systems are data-driven models that successfully pro- vide users with personalized rankings of items (movies, books...). Meanwhile, for user minority groups, those systems can be unfair in predicting users’ expectations due to biased data. Consequently, fairness remains an open challenge in the rank- ing prediction task. To address this issue, we propose in this paper FairBayRank, a fair Bayesian personalized ranking algorithm that deals with both fairness and ranking performance requirements. FairBayRank evaluation on real-world datasets shows that it efficiently alleviates unfairness issues while ensuring high prediction performances.

Original language	English
Title of host publication	31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning Bruges, Belgium October 04 - 06
Publication status	Published - 2023
Event	31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning - Bruges, Belgium Duration: 4 Oct 2023 → 6 Oct 2023 Conference number: 31 https://www.esann.org

Conference

Conference	31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Country/Territory	Belgium
City	Bruges
Period	4/10/23 → 6/10/23
Internet address	https://www.esann.org

Access to Document

ES2023-81Licence: Other

https://www.esann.org/sites/default/files/proceedings/2023/ES2023-81.pdfLicence: Other

Cite this

@inproceedings{f422027f152a43e795509a5ea19193a6,

title = "FairBayRank: A Fair Personalized Bayesian Ranker",

abstract = "Recommender systems are data-driven models that successfully pro- vide users with personalized rankings of items (movies, books...). Meanwhile, for user minority groups, those systems can be unfair in predicting users{\textquoteright} expectations due to biased data. Consequently, fairness remains an open challenge in the rank- ing prediction task. To address this issue, we propose in this paper FairBayRank, a fair Bayesian personalized ranking algorithm that deals with both fairness and ranking performance requirements. FairBayRank evaluation on real-world datasets shows that it efficiently alleviates unfairness issues while ensuring high prediction performances.",

author = "{Noulapeu Ngaffo}, Armielle and Julien Albert and Beno{\^i}t Fr{\'e}nay and Gilles Perrouin",

year = "2023",

language = "English",

booktitle = "31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning Bruges, Belgium October 04 - 06",

note = "31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning ; Conference date: 04-10-2023 Through 06-10-2023",

url = "https://www.esann.org",

}

Noulapeu Ngaffo, A , Albert, J , Frénay, B & Perrouin, G 2023, FairBayRank: A Fair Personalized Bayesian Ranker. in 31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning Bruges, Belgium October 04 - 06. 31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, Belgium, 4/10/23. <https://www.esann.org/sites/default/files/proceedings/2023/ES2023-81.pdf>

TY - GEN

T1 - FairBayRank

T2 - 31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

AU - Noulapeu Ngaffo, Armielle

AU - Albert, Julien

AU - Frénay, Benoît

AU - Perrouin, Gilles

N1 - Conference code: 31

PY - 2023

Y1 - 2023

N2 - Recommender systems are data-driven models that successfully pro- vide users with personalized rankings of items (movies, books...). Meanwhile, for user minority groups, those systems can be unfair in predicting users’ expectations due to biased data. Consequently, fairness remains an open challenge in the rank- ing prediction task. To address this issue, we propose in this paper FairBayRank, a fair Bayesian personalized ranking algorithm that deals with both fairness and ranking performance requirements. FairBayRank evaluation on real-world datasets shows that it efficiently alleviates unfairness issues while ensuring high prediction performances.

AB - Recommender systems are data-driven models that successfully pro- vide users with personalized rankings of items (movies, books...). Meanwhile, for user minority groups, those systems can be unfair in predicting users’ expectations due to biased data. Consequently, fairness remains an open challenge in the rank- ing prediction task. To address this issue, we propose in this paper FairBayRank, a fair Bayesian personalized ranking algorithm that deals with both fairness and ranking performance requirements. FairBayRank evaluation on real-world datasets shows that it efficiently alleviates unfairness issues while ensuring high prediction performances.

M3 - Conference contribution

BT - 31st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning Bruges, Belgium October 04 - 06

Y2 - 4 October 2023 through 6 October 2023

ER -

FairBayRank: A Fair Personalized Bayesian Ranker

Abstract

Conference

Access to Document

Fingerprint

Cite this