Controlling Popularity Bias in Learning-to-Rank Recommendation

Abdollahpouri, Himan; Burke, Robin; Mobasher, Bamshad

doi:10.1145/3109859.3109912

Cited by 317 publications

(270 citation statements)

References 8 publications

Supporting

Mentioning

251

Contrasting

Unclassified

Order By: Relevance

“…Recommender systems are quintessential tools to support users in finding relevant information in large information spaces [10]. However, one limitation of typical recommender systems is the so-called popularity bias, which leads to the underrepresentation of less popular (i.e., long-tail) items in the recommendation lists [1,4,5]. The recent work of Abdollahpouri et al [2] has investigated this popularity bias from the user perspective in the movie domain.…”

Section: Introductionmentioning

confidence: 99%

The Unfairness of Popularity Bias in Music Recommendation: A Reproducibility Study

Kowald

Schedl

Lex

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Research has shown that recommender systems are typically biased towards popular items, which leads to less popular items being underrepresented in recommendations. The recent work of Abdollahpouri et al. in the context of movie recommendations has shown that this popularity bias leads to unfair treatment of both long-tail items as well as users with little interest in popular items. In this paper, we reproduce the analyses of Abdollahpouri et al. in the context of music recommendation. Specifically, we investigate three user groups from the Last.fm music platform that are categorized based on how much their listening preferences deviate from the most popular music among all Last.fm users in the dataset: (i) low-mainstream users, (ii) medium-mainstream users, and (iii) high-mainstream users. In line with Abdollahpouri et al., we find that state-of-the-art recommendation algorithms favor popular items also in the music domain. However, their proposed Group Average Popularity metric yields different results for Last.fm than for the movie domain, presumably due to the larger number of available items (i.e., music artists) in the Last.fm dataset we use. Finally, we compare the accuracy results of the recommendation algorithms for the three user groups and find that the low-mainstreaminess group significantly receives the worst recommendations.

show abstract

Section: Introductionmentioning

confidence: 99%

The Unfairness of Popularity Bias in Music Recommendation: A Reproducibility Study

Kowald

Schedl

Lex

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Controlling popularity bias is important for many recommender systems as it affects the fairness of the platform and it also improves the novelty of the recommendations. For more information on popularity bias see [1,3]main. We wanted to go a bit deep in this problem and find different types of solutions for it to make a better sense of this type of objective and to be able to generalize for similar problems.…”

Section: Results and The Progress So Farmentioning

confidence: 99%

“…There is a large body of recent work on incorporating diversity, novelty, long tail promotion and other metrics as additional objectives for recommendation generation and evaluation. See, for example, [3,16]. There is also a growing body of work on combining multiple objectives using constraint optimization techniques, including linear programming.…”

Section: State Of the Artmentioning

confidence: 99%

Companion Proceedings of The 2019 World Wide Web Conference

2019

View full text Add to dashboard Cite

One of the most essential parts of any recommender system is personalization-how acceptable the recommendations are from the user's perspective. However, in many real-world applications, there are other stakeholders whose needs and interests should be taken into account. In this work, we define the problem of multistakeholder recommendation and we focus on finding algorithms for a special case where the recommender system itself is also a stakeholder. In addition, we will explore the idea of incremental incorporation of system-level objectives into recommender systems over time to tackle the existing problems in the optimization techniques which only look for optimizing the individual users' lists. CCS CONCEPTS• Computing methodologies → Machine learning.

show abstract

“…Biega et al [9] considered individual producer fairness in ranking in gig-economy platforms. Kamishima et al [32] and Abdollahpouri et al [2] reduced popularity bias among producers while Patro et al [41] addressed fairness issues arising due to frequent updates of platforms. However, these papers did not study the trade-off between producer and customer fairness, and the cost of achieving one over the other.…”

Section: Background and Related Workmentioning

confidence: 99%

FairRec: Two-Sided Fairness for Personalized Recommendations in Two-Sided Platforms

Patro

Biswas

Ganguly

et al. 2020

Proceedings of the Web Conference 2020

148

View full text Add to dashboard Cite

We investigate the problem of fair recommendation in the context of two-sided online platforms, comprising customers on one side and producers on the other. Traditionally, recommendation services in these platforms have focused on maximizing customer satisfaction by tailoring the results according to the personalized preferences of individual customers. However, our investigation reveals that such customer-centric design may lead to unfair distribution of exposure among the producers, which may adversely impact their well-being. On the other hand, a producer-centric design might become unfair to the customers. Thus, we consider fairness issues that span both customers and producers. Our approach involves a novel mapping of the fair recommendation problem to a constrained version of the problem of fairly allocating indivisible goods. Our proposed FairRec algorithm guarantees at least Maximin Share (MMS) of exposure for most of the producers and Envy-Free up to One item (EF1) fairness for every customer. Extensive evaluations over multiple real-world datasets show the effectiveness of FairRec in ensuring two-sided fairness while incurring a marginal loss in the overall recommendation quality. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

Controlling Popularity Bias in Learning-to-Rank Recommendation

Cited by 317 publications

References 8 publications

The Unfairness of Popularity Bias in Music Recommendation: A Reproducibility Study

The Unfairness of Popularity Bias in Music Recommendation: A Reproducibility Study

Companion Proceedings of The 2019 World Wide Web Conference

FairRec: Two-Sided Fairness for Personalized Recommendations in Two-Sided Platforms

Contact Info

Product

Resources

About