Unmasking Gender Bias in Recommendation Systems and Enhancing Category-Aware Fairness

• URL: http://arxiv.org/abs/2502.17921v1
• Date: Tue, 25 Feb 2025 07:37:28 GMT
• Title: Unmasking Gender Bias in Recommendation Systems and Enhancing Category-Aware Fairness
• Authors: Tahsin Alamgir Kheya, Mohamed Reda Bouadjenek, Sunil Aryal,
• Abstract summary: We introduce a set of comprehensive metrics for gender bias in recommendations.<n>We show the importance of evaluating fairness on a more granular level.<n>We show that employing a category-aware fairness metric as a regularization term along with the main recommendation loss during training can help effectively minimize bias in the models' output.
• Score: 2.124791625488617
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Recommendation systems are now an integral part of our daily lives. We rely on them for tasks such as discovering new movies, finding friends on social media, and connecting job seekers with relevant opportunities. Given their vital role, we must ensure these recommendations are free from societal stereotypes. Therefore, evaluating and addressing such biases in recommendation systems is crucial. Previous work evaluating the fairness of recommended items fails to capture certain nuances as they mainly focus on comparing performance metrics for different sensitive groups. In this paper, we introduce a set of comprehensive metrics for quantifying gender bias in recommendations. Specifically, we show the importance of evaluating fairness on a more granular level, which can be achieved using our metrics to capture gender bias using categories of recommended items like genres for movies. Furthermore, we show that employing a category-aware fairness metric as a regularization term along with the main recommendation loss during training can help effectively minimize bias in the models' output. We experiment on three real-world datasets, using five baseline models alongside two popular fairness-aware models, to show the effectiveness of our metrics in evaluating gender bias. Our metrics help provide an enhanced insight into bias in recommended items compared to previous metrics. Additionally, our results demonstrate how incorporating our regularization term significantly improves the fairness in recommendations for different categories without substantial degradation in overall recommendation performance.

Related papers

• The Root Shapes the Fruit: On the Persistence of Gender-Exclusive Harms in Aligned Language Models [58.130894823145205]
  We center transgender, nonbinary, and other gender-diverse identities to investigate how alignment procedures interact with pre-existing gender-diverse bias. Our findings reveal that DPO-aligned models are particularly sensitive to supervised finetuning. We conclude with recommendations tailored to DPO and broader alignment practices.
  arXiv  Detail & Related papers  (2024-11-06T06:50:50Z)
• GenderCARE: A Comprehensive Framework for Assessing and Reducing Gender Bias in Large Language Models [73.23743278545321]
  Large language models (LLMs) have exhibited remarkable capabilities in natural language generation, but have also been observed to magnify societal biases. GenderCARE is a comprehensive framework that encompasses innovative Criteria, bias Assessment, Reduction techniques, and Evaluation metrics.
  arXiv  Detail & Related papers  (2024-08-22T15:35:46Z)
• A Survey on Fairness-aware Recommender Systems [59.23208133653637]
  We present concepts of fairness in different recommendation scenarios, comprehensively categorize current advances, and introduce typical methods to promote fairness in different stages of recommender systems. Next, we delve into the significant influence that fairness-aware recommender systems exert on real-world industrial applications.
  arXiv  Detail & Related papers  (2023-06-01T07:08:22Z)
• Gender Biases in Automatic Evaluation Metrics for Image Captioning [87.15170977240643]
  We conduct a systematic study of gender biases in model-based evaluation metrics for image captioning tasks. We demonstrate the negative consequences of using these biased metrics, including the inability to differentiate between biased and unbiased generations. We present a simple and effective way to mitigate the metric bias without hurting the correlations with human judgments.
  arXiv  Detail & Related papers  (2023-05-24T04:27:40Z)
• Improving Recommendation Fairness via Data Augmentation [66.4071365614835]
  Collaborative filtering based recommendation learns users' preferences from all users' historical behavior data, and has been popular to facilitate decision making. A recommender system is considered unfair when it does not perform equally well for different user groups according to users' sensitive attributes. In this paper, we study how to improve recommendation fairness from the data augmentation perspective.
  arXiv  Detail & Related papers  (2023-02-13T13:11:46Z)
• Equal Experience in Recommender Systems [21.298427869586686]
  We introduce a novel fairness notion (that we call equal experience) to regulate unfairness in the presence of biased data. We propose an optimization framework that incorporates the fairness notion as a regularization term, as well as introduce computationally-efficient algorithms that solve the optimization.
  arXiv  Detail & Related papers  (2022-10-12T05:53:05Z)
• Experiments on Generalizability of User-Oriented Fairness in Recommender Systems [2.0932879442844476]
  A fairness-aware recommender system aims to treat different user groups similarly. We propose a user-centered fairness re-ranking framework applied on top of a base ranking model. We evaluate the final recommendations provided by the re-ranking framework from both user- (e.g., NDCG) and item-side (e.g., novelty, item-fairness) metrics.
  arXiv  Detail & Related papers  (2022-05-17T12:36:30Z)
• Unbiased Pairwise Learning to Rank in Recommender Systems [4.058828240864671]
  Unbiased learning to rank algorithms are appealing candidates and have already been applied in many applications with single categorical labels. We propose a novel unbiased LTR algorithm to tackle the challenges, which innovatively models position bias in the pairwise fashion. Experiment results on public benchmark datasets and internal live traffic show the superior results of the proposed method for both categorical and continuous labels.
  arXiv  Detail & Related papers  (2021-11-25T06:04:59Z)
• Revisiting Popularity and Demographic Biases in Recommender Evaluation and Effectiveness [6.210698627561645]
  We investigate how recommender performance varies according to popularity and demographics. We find statistically significant differences in recommender performance by both age and gender. We observe that recommendation utility steadily degrades for older users, and is lower for women than men.
  arXiv  Detail & Related papers  (2021-10-15T20:30:51Z)
• Fairness-Aware Explainable Recommendation over Knowledge Graphs [73.81994676695346]
  We analyze different groups of users according to their level of activity, and find that bias exists in recommendation performance between different groups. We show that inactive users may be more susceptible to receiving unsatisfactory recommendations, due to insufficient training data for the inactive users. We propose a fairness constrained approach via re-ranking to mitigate this problem in the context of explainable recommendation over knowledge graphs.
  arXiv  Detail & Related papers  (2020-06-03T05:04:38Z)
• Opportunistic Multi-aspect Fairness through Personalized Re-ranking [5.8562079474220665]
  We present a re-ranking approach to fairness-aware recommendation that learns individual preferences across multiple fairness dimensions. We show that our opportunistic and metric-agnostic approach achieves a better trade-off between accuracy and fairness than prior re-ranking approaches.
  arXiv  Detail & Related papers  (2020-05-21T04:25:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.