Designing and Evaluating Calibration-Driven Recommender Systems for Fairness and Explainability

• Paul Atauchi Universidade de São Paulo (USP)
• Andre Zanon Universidade de São Paulo (USP)
• Leonardo Rocha Universidade Federal de São Jõao del-Rei (UFSJ)
• Marcelo Manzato Universidade de São Paulo (USP)

DOI: https://doi.org/10.5753/sbbd.2026.249259

Resumo

Recommender systems are increasingly required to satisfy fairness and transparency constraints. However, the interaction between calibration-based fairness interventions and recommendation explainability remains insufficiently understood. This paper proposes a calibration-driven framework that extends single-category calibration to a multi-category setting, enabling a richer representation of user preferences across item dimensions. The approach integrates fairness-aware calibration with post-hoc knowledge-graph-based explanations into a unified pipeline. We evaluate both recommendation performance and explanation quality, considering alignment with user profiles, accuracy, miscalibration, and popularity bias, as well as explanation metrics such as recency, popularity, and diversity. Results show improved alignment, reduced miscalibration and popularity bias, and preserved accuracy, while explanation quality remains stable. These findings indicate that calibration-based fairness can be achieved without compromising explainability.

Palavras-chave: Recommender Systems, Fairness-Aware Recommendation, Multi-Category Calibration, Explainable Artificial Intelligence (XAI), Knowledge Graph-Based Explanations

Referências

Abdollahpouri, H., Mansoury, M., Burke, R., and Mobasher, B. (2019). The unfairness of popularity bias in recommendation. arXiv preprint arXiv:1907.13286.

Abdollahpouri, H., Mansoury, M., Burke, R., Mobasher, B., and Malthouse, E. (2021). User-centered evaluation of popularity bias in recommender systems. In Proc. of the 29th ACM conference on user modeling, adaptation and personalization.

Aggarwal, C. C. et al. (2016). Recommender systems, volume 1. Springer.

Arévalo, I. and Salmeron, J. L. (2025). Model-agnostic post-hoc explainability for recommender systems. Expert Systems with Applications, page 129608.

Atauchi, P., Zanon, A., Rocha, L., and Manzato, M. (2025a). Aprimorando a personalização de sistemas de recomendação por calibração multicategoria. In Proceedings of the 31st WebMedia.

Atauchi, P. D. F., Zanon, A. L., Rocha, L. C. D. d., and Manzato, M. G. (2025b). Do calibrated recommendations affect explanations? a study on post-hoc adjustments. Journal on Interactive Systems, 16(1):441–460.

Balloccu, G., Boratto, L., Fenu, G., and Marras, M. (2022). Post processing recommender systems with knowledge graphs for recency, popularity, and diversity of explanations. In Proceedings of the 45th International ACM SIGIR, SIGIR ’22.

Cantador, I., Brusilovsky, P., and Kuflik, T. (2011). 2nd workshop on information heterogeneity and fusion in recommender systems (hetrec 2011). In Proceedings of the 5th ACM conference on Recommender systems, RecSys 2011.

da Silva, D. C. and Durão, F. A. (2023). Introducing a framework and a decision protocol to calibrated recommender systems: Dc da silva, fa durão. Applied Intelligence, 53(19):22044–22072.

da Silva, D. C., Manzato, M. G., and Durão, F. A. (2021). Exploiting personalized calibration and metrics for fairness recommendation. Expert Systems with Applications, 181:115112.

da Silva, D. C., Pires, J. M., and Durão, F. A. (2025). Exploiting surrogate submodular and cost-effective lazy forward algorithms for calibrated recommendations. In Simpósio Brasileiro de Banco de Dados (SBBD), pages 98–111. SBC.

Du, Y., Ranwez, S., Sutton-Charani, N., and Ranwez, V. (2022). Post-hoc recommendation explanations through an efficient exploitation of the dbpedia category hierarchy. Knowledge-Based Systems, 245:108560.

Ge, Y., Liu, S., Fu, Z., Tan, J., Li, Z., Xu, S., Li, Y., Xian, Y., and Zhang, Y. (2024). A survey on trustworthy recommender systems. ACM TORS, 3(2):1–68.

Harper, F. M. and Konstan, J. A. (2015). The movielens datasets: History and context. Acm transactions on interactive intelligent systems (tiis), 5(4):1–19.

He, X., Liao, L., Zhang, H., Nie, L., Hu, X., and Chua, T.-S. (2017). Neural collaborative filtering. In Proceedings of the 26th International Conference on World Wide Web.

Lima, R. O., Comarella, G., Belém, F., and Reis, J. C. (2022). Justiça em sistemas de recomendação: Uma análise de técnicas de regularização. In SBBD.

Liu, X., Yu, T., Xie, K., Wu, J., and Li, S. (2024). Interact with the explanations: Causal debiased explainable recommendation system. In Proceedings of the 17th ACM International Conference on Web Search and Data Mining, pages 472–481.

Musto, C., Narducci, F., Lops, P., De Gemmis, M., and Semeraro, G. (2016). Explod: A framework for explaining recommendations based on the linked open data cloud. In Proceedings of the 10th ACM Conference on Recommender Systems, pages 151–154.

Musto, C., Narducci, F., Lops, P., De Gemmis, M., and Semeraro, G. (2019). Linked open data-based explanations for transparent recommender systems. International Journal of Human-Computer Studies, 121:93–107.

Naghiaei, M., Dehghan, M., Rahmani, H. A., Azizi, J., and Aliannejadi, M. (2024). Personalized beyond-accuracy calibration in recommendation. In Proceedings of the 2024 ACM SIGIR.

Palomino, Y. Q., Souza, R. F., Atauchi, P. D. F., and Manzato, M. G. (2026). Calibration of personality-based recommender systems: Balancing relevance and diversity. In Proceedings of the 41st ACM/SIGAPP Symposium on Applied Computing (SAC ’26).

Ping, Y., Li, Y., and Zhu, J. (2025). Beyond accuracy measures: the effect of diversity, novelty and serendipity in recommender systems on user engagement. Electronic Commerce Research, 25(3):2177–2204.

Rendle, S., Freudenthaler, C., Gantner, Z., and Schmidt-Thieme, L. (2009). Bpr: Bayesian personalized ranking from implicit feedback. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, UAI ’09.

Roy, D. and Dutta, M. (2022). A systematic review and research perspective on recommender systems. Journal of Big Data, 9(1):59.

Sacilotti, A., Souza, R. F. d., and Manzato, M. G. (2023). Counteracting popularity-bias and improving diversity through calibrated recommendations. In Proceedings.

Seabra, A., Cavalcante, C., and Lifschitz, S. (2025). Towards a self-explainable recommender system using large language models and knowledge graphs. In Simpósio Brasileiro de Banco de Dados (SBBD), pages 479–489. SBC.

Souza, R. and Manzato, M. (2024a). Explorando formas de calibração e redução do viés de popularidade em sistemas de recomendação. In Anais Estendidos do XXX Simpósio Brasileiro de Sistemas Multimídia e Web, pages 9–10, Porto Alegre, RS, Brasil. SBC.

Souza, R. and Manzato, M. (2024b). A two-stage calibration approach for mitigating bias and fairness in recommender systems. In Proceedings of the 39th ACM SAC.

Steck, H. (2018). Calibrated recommendations. In Proceedings of the 12th ACM conference on recommender systems, pages 154–162.

Zhang, Y. and Chen, X. (2020). Explainable recommendation: A survey and new perspectives. Foundations and Trends® in Information Retrieval, 14(1):1–101.

Zhao, Y., Wang, Y., Liu, Y., Cheng, X., Aggarwal, C. C., and Derr, T. (2025). Fairness and diversity in recommender systems: a survey. ACM Transactions on Intelligent Systems and Technology, 16(1):1–28.