Group Fairness in Recommendation Systems: The Importance of Hierarchical Clustering in Identifying Latent Groups in MovieLens and Amazon Books

Rafael Vargas Mesquita dos Santos; Giovanni Ventorim Comarela

doi:10.5753/jis.2025.5407

Authors

Rafael Vargas Mesquita dos Santos Federal Institute of Espírito Santo https://orcid.org/0009-0003-2915-7642
Giovanni Ventorim Comarela Federal University of Espírito Santo https://orcid.org/0000-0001-7612-9650

DOI:

https://doi.org/10.5753/jis.2025.5407

Keywords:

Recommendation System, Group Fairness, Agglomerative Hierarchical Clustering, Latent Groups

Abstract

Fairness in recommendation systems is a critical area of study, particularly when addressing group disparities based on sensitive attributes such as gender, age, activity levels, or user location. This study also explores latent groups identified through hierarchical clustering techniques. The goal is to assess group unfairness across various clustering configurations and collaborative filtering strategies to promote equitable and inclusive recommendation systems. We applied collaborative filtering techniques, including ALS, KNN, and NMF, and evaluated group unfairness using metrics such as R_grp for different clustering configurations (e.g., gender, age, activity level, location, and hierarchical clustering) in two datasets: MovieLens and Amazon Books. Hierarchical clustering yielded the highest group unfairness, with ALS and NMF reaching R_grp values of 0.0062 and 0.0049 in MovieLens, and NMF and KNN peaking at 0.0972 and 0.0220 in Amazon Books. These results reveal significant fairness disparities across both latent and observable user groups, reinforcing the importance of selecting appropriate filtering strategies and clustering methods to build fair and inclusive recommendation systems.

Downloads

Download data is not yet available.

References

Abdollahpouri, H., Adomavicius, G., Burke, R., Guy, I., Jannach, D., Kamishima, T., Krasnodebski, J., and Pizzato, L. (2020). Multistakeholder recommendation: Survey and research directions. User Modeling and User-Adapted Interaction, 30. DOI: https://doi.org/10.1007/s11257-019-09256-1.

Alves, G., Jannach, D., Ferrari De Souza, R., and Manzato, M. G. (2024a). User perception of fairness-calibrated recommendations. In Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization, UMAP ’24, pp. 78–88, New York, NY, USA. Association for Computing Machinery. DOI: https://doi.org/10.1145/3627043.3659558.

Alves, P., Martins, A., Negrão, F., Novais, P., Almeida, A., and Marreiros, G. (2024b). Are heterogeinity and conflicting preferences no longer a problem? Personality-based dynamic clustering for group recommender systems. Expert Systems with Applications, 255:124812. DOI: https://doi.org/10.1016/j.eswa.2024.124812.

Bagchi, S. (2022). Books Dataset: Books, Users and Ratings. Kaggle.

Bernardino, G. S. and Gonçalves, A. L. (2019). A education profile model applied in the context of recommender systems. IEEE Latin America Transactions, 17(3):505–512. DOI: https://doi.org/10.1109/TLA.2019.8863321.

Beutel, A., Chen, J., Doshi, T., Qian, H., Wei, L., Wu, Y., Heldt, L., Zhao, Z., Hong, L., Chi, E. H., and Goodrow, C. (2019). Fairness in recommendation ranking through pairwise comparisons. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 2212–2220. DOI: https://doi.org/10.1145/3292500.3330745.

Burke, R., Sonboli, N., and Ordonez-Gauger, A. (2018). Balanced neighborhoods for multi-sided fairness in recommendation. In Friedler, S. A. and Wilson, C. (Eds.), Proceedings of the 1st Conference on Fairness, Accountability and Transparency, vol. 81 of Proceedings of Machine Learning Research, pp. 202–214. PMLR.

Cavalcante, C. G. S. and Fettermann, D. C. (2019). Recommendations for product development of intelligent products. IEEE Latin America Transactions, 17(10):1645–1652. DOI: https://doi.org/10.1109/TLA.2019.8986442.

Chen, J., Feng, N., Bernstein, M. S., and Zhu, H. (2018). Investigating the impacts of gender representation in recommendation. In Workshop on Recommendation in Multistakeholder Environments at the ACM RecSys.

Coelho, N. L. L., Figueiredo, T. F., and Figueiredo, R. (2023). Uso de programação linear inteira para geração e análise de agrupamentos de políticos da câmara dos deputados. In BraSNAM 2023 - XII Brazilian Workshop on Social Network Analysis and Mining, João Pessoa, PB, Brasil. Sociedade Brasileira de Computação. DOI: https://doi.org/10.5753/brasnam.2023.230858.

Crenshaw, K. (1989). Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 1989(1):139–167. DOI: https://doi.org/10.4324/9780429499142-5.

Deldjoo, Y., Jannach, D., Bellogin, A., Difonzo, A., and Zanzonelli, D. (2024). Fairness in recommender systems: Research landscape and future directions. User Modeling and User-Adapted Interaction, 34(1):59–108. DOI: https://doi.org/10.1007/s11257-023-09364-z.

Dwork, C., Hardt, M., Pitassi, T., Reingold, O., and Zemel, R. (2011). Fairness through awareness.

Ekstrand, M. D., Das, A., Burke, R., and Diaz, F. (2022). Fairness in information access systems. Foundations and Trends® in Information Retrieval, 16(1–2):1–177. DOI: https://doi.org/10.1561/1500000079.

Ekstrand, M. D., Tian, M., Azpiazu, I. M., Ekstrand, J. D., Anuyah, O., McNeill, D., and Pera, M. S. (2018). All the cool kids, how do they fit in?: Popularity and demographic biases in recommender evaluation and effectiveness. In Friedler, S. A. and Wilson, C. (Eds.), Proceedings of the 1st Conference on Fairness, Accountability and Transparency, vol. 81 of Proceedings of Machine Learning Research, pp. 172–186. PMLR.

Friedler, S. A., Scheidegger, C., and Venkatasubramanian, S. (2016). On the (im)possibility of fairness.

Harper, F. M. and Konstan, J. A. (2015). The MovieLens datasets: History and context. ACM Transactions on Interactive Intelligent Systems (TiiS), 5(4):Article 19. DOI: https://doi.org/10.1145/2827872.

Hazrati, N. and Ricci, F. (2024). Choice models and recommender systems effects on users’ choices. User Model User-Adap Inter, 34(1):109–145. DOI: https://doi.org/10.1007/s11257-023-09366-x.

Jáñez-Martino, F., Alaiz-Rodríguez, R., González-Castro, V., Fidalgo, E., and Alegre, E. (2023). Classifying spam emails using agglomerative hierarchical clustering and a topic-based approach. Applied Soft Computing, 139:110226. DOI: https://doi.org/10.1016/j.asoc.2023.110226.

Leonhardt, J., Anand, A., and Khosla, M. (2018). User fairness in recommender systems. CoRR, abs/1807.06349. DOI: https://doi.org/10.1145/3184558.3186949.

Li, Y., Chen, H., Xu, S., Ge, Y., Tan, J., Liu, S., and Zhang, Y. (2023). Fairness in recommendation: Foundations, methods and applications.

Li, Y., Chen, H., Xu, S., Ge, Y., and Zhang, Y. (2021). Towards personalized fairness based on causal notion. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’21, pp. 1054–1063, New York, NY, USA. Association for Computing Machinery. DOI: https://doi.org/10.1145/3404835.3462966.

Liu, S., Ge, Y., Xu, S., Zhang, Y., and Marian, A. (2022). Fairness-aware federated matrix factorization. In Proceedings of the 16th ACM Conference on Recommender Systems, RecSys ’22, pp. 168–178, New York, NY, USA. Association for Computing Machinery. DOI: https://doi.org/10.1145/3523227.3546771.

Liu, S., Sun, J., Deng, X., and et al. (2024). Towards platform profit-aware fairness in personalized recommendation. International Journal of Machine Learning and Cybernetics. DOI: https://doi.org/10.1007/s13042-024-02149-9.

Pereira, F. S. F., Linhares, C. D. G., Ponciano, J. R., Gama, J., de Amo, S., and Oliveira, G. M. B. (2018). That’s my jam! Uma análise temporal sobre a evolução das preferências dos usuários em uma rede social de músicas. In BraSNAM 2018 - VII Brazilian Workshop on Social Network Analysis and Mining, Porto Alegre, RS, Brasil. Sociedade Brasileira de Computação. DOI: https://doi.org/10.5753/brasnam.2018.3587.

Rahmani, H. A., Deldjoo, Y., Tourani, A., and Naghiaei, M. (2022). The unfairness of active users and popularity bias in point-of-interest recommendation.

Rastegarpanah, B., Gummadi, K. P., and Crovella, M. (2019). Fighting fire with fire: Using antidote data to improve polarization and fairness of recommender systems. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, WSDM ’19. ACM. DOI: https://doi.org/10.1145/3289600.3291002.

Santos, R. and Comarela, G. (2024). Development of an equity strategy for recommendation systems. In Anais do V Workshop sobre as Implicações da Computação na Sociedade, pp. 24–35, Porto Alegre, RS, Brasil. SBC. DOI: https://doi.org/10.5753/wics.2024.1975.

Santos, R. V. M., Comarela, G. V., and Junior, M. C. G. L. (2024). Implementation of fairness measures: A case study in the cultural context for different strategies in recommendation systems. iSys - Revista Brasileira de Sistemas de Informação, 17:1–16. DOI: https://doi.org/10.5753/isys.2024.4214.

Sonboli, N., Eskandanian, F., Burke, R., Liu, W., and Mobasher, B. (2020). Opportunistic multiaspect fairness through personalized re-ranking. UMAP ’20, pp. 239–247, New York, NY, USA. Association for Computing Machinery. DOI: https://doi.org/10.1145/3340631.3394846.

Sonboli, N., Smith, J. J., Cabral Berenfus, F., Burke, R., and Fiesler, C. (2021). Fairness and transparency in recommendation: The users’ perspective. In Proceedings of the 29th ACM Conference on User Modeling, Adaptation and Personalization, UMAP ’21, pp. 274–279. ACM. DOI: https://doi.org/10.1145/3450613.3456835.

Souza, E., Lichtnow, D., and Gasparini, I. (2022). Estratégia de pós-processamento aplicada a um sistema de recomendação de artigos para a melhora da diversidade. In BraSNAM 2022 - XI Brazilian Workshop on Social Network Analysis and Mining, pp. 216–221, Porto Alegre, RS, Brasil. Sociedade Brasileira de Computação. DOI: https://doi.org/10.5753/brasnam.2022.222805.

Wang, Y., Ma, W., Zhang, M., Liu, Y., and Ma, S. (2023). A survey on the fairness of recommender systems. ACM Transactions on Information Systems, 41(3). DOI: https://doi.org/10.1145/3547333.

Wu, Y., Cao, J., Xu, G., and Tan, Y. (2021). Tfrom: A two-sided fairness-aware recommendation model for both customers and providers.

Zafar, M., Valera, I., Rodriguez, M., and Gummadi, K. P. (2017). Fairness beyond disparate treatment & disparate impact: Learning classification without disparate mistreatment. In WWW ’17: Proceedings of the 26th International Conference on World Wide Web, pp. 1171–1180. DOI: https://doi.org/10.1145/3038912.3052660.

Zehlike, M., Yang, K., and Stoyanovich, J. (2022). Fairness in ranking: A survey. ACM Computing Surveys, 55(6):1–34. DOI: https://doi.org/10.1145/3533379.