Aprimorando a Personalização de Sistemas de Recomendação por Calibração Multicategoria

• Paul Atauchi USP
• André Levi Zanon University College Cork
• Leonardo Rocha UFSJ
• Marcelo G. Manzato USP

DOI: https://doi.org/10.5753/webmedia.2025.16145

Resumo

Recommender systems (RSs) are widely used to personalize suggestions based on users’ historical data. Despite recent advances, many RSs still prioritize accuracy metrics, often generating nonpersonalized lists that fail to reflect users’ actual interests. Calibration strategies have been proposed to address this misalignment by tailoring recommendations to users’ profiles. However, most approaches consider only a single category (e.g., genre), limiting their ability to capture the multifaceted nature of user preferences. This paper proposes a multi-category calibration method that simultaneously incorporates multiple categories extracted from knowledge graphs. The approach aims to balance recommendation relevance with alignment across diverse user profile dimensions. We evaluate the method on two well-known datasets, MovieLens and LastFM, using two contrasting recommendation algorithms: NCF (neural) and BPR-MF (non-neural). Results show that multicategory calibration can improve alignment with multifaceted user preferences while maintaining or enhancing accuracy, fostering fairer, more personalized, and more relevant recommendations.

Palavras-chave: Recomendação, Calibração multicategoria, Personalização

Referências

Himan Abdollahpouri, Masoud Mansoury, Robin Burke, and Bamshad Mobasher. 2020. The connection between popularity bias, calibration, and fairness in recommendation. In Proceedings of the 14th ACM conference on recommender systems. 726–731.

Himan Abdollahpouri, Zahra Nazari, Alex Gain, Clay Gibson, Maria Dimakopoulou, Jesse Anderton, Benjamin Carterette, Mounia Lalmas, and Tony Jebara. 2023. Calibrated recommendations as a minimum-cost flow problem. In Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining. 571–579.

Iván Cantador, Peter Brusilovsky, and Tsvi Kuflik. 2011. 2nd Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2011). In Proceedings of the 5th ACM conference on Recommender systems (Chicago, IL, USA) (RecSys 2011). ACM, New York, NY, USA.

Diego Corrêa da Silva and Frederico Araújo Durão. 2025. Benchmarking fairness measures for calibrated recommendation systems on movies domain. Expert Systems with Applications 269 (2025), 126380.

Diego Corrêa da Silva, Marcelo Garcia Manzato, and Frederico Araújo Durão. 2021. Exploiting personalized calibration and metrics for fairness recommendation. Expert Systems with Applications 181 (2021), 115112.

Diego Corrêa da Silva, Marcelo Garcia Manzato, and Frederico Araújo Durão. 2021. Exploiting personalized calibration and metrics for fairness recommendation. Expert Systems with Applications 181 (2021), 115112. DOI: 10.1016/j.eswa.2021.115112

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

Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, and Tat-Seng Chua. 2017. Neural Collaborative Filtering. In Proceedings of the 26th International Conference onWorld WideWeb (Perth, Australia) (WWW’17). InternationalWorld Wide Web Conferences Steering Committee, Republic and Canton of Geneva, CHE, 173–182. DOI: 10.1145/3038912.3052569

Mohammadmehdi Naghiaei, Mahdi Dehghan, Hossein A Rahmani, Javad Azizi, and Mohammad Aliannejadi. 2024. Personalized beyond-accuracy calibration in recommendation. In Proceedings of the 2024 ACM SIGIR International Conference on Theory of Information Retrieval. 107–116.

Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme. 2009. BPR: Bayesian Personalized Ranking from Implicit Feedback. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence (Montreal, Quebec, Canada) (UAI ’09). AUAI Press, Arlington, Virginia, USA, 452–461.

Francesco Ricci, Lior Rokach, and Bracha Shapira. 2022. Recommender Systems: Techniques, Applications, and Challenges. Springer US, New York, NY, 1–35. DOI: 10.1007/978-1-0716-2197-4_1

Andre Sacilotti, Rodrigo Ferrari de Souza, and Marcelo Garcia Manzato. 2023. Counteracting popularity-bias and improving diversity through calibrated recommendations. In Proceedings.

Diego Silva and Frederico Durão. 2022. Explorando Justiça em Sistemas de Recomendação. In Anais Estendidos do XXVIII Simpósio Brasileiro de Sistemas Multimídia e Web (Curitiba). SBC, Porto Alegre, RS, Brasil, 11–14. DOI: 10.5753/webmedia_estendido.2022.225303

Rodrigo Souza and Marcelo Manzato. 2024. 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 (Juiz de Fora/MG). SBC, Porto Alegre, RS, Brasil, 9–10. DOI: 10.5753/webmedia_estendido.2024.244380

Harald Steck. 2018. Calibrated recommendations. In Proceedings of the 12th ACM conference on recommender systems. 154–162.

YifanWang,Weizhi Ma, Min Zhang, Yiqun Liu, and Shaoping Ma. 2023. A survey on the fairness of recommender systems. ACM Transactions on Information Systems 41, 3 (2023), 1–43.

André Zanon, Leonardo Rocha, and Marcelo Manzato. 2024. O Impacto de Estratégias de Embeddings de Grafos na Explicabilidade de Sistemas de Recomendação. In Proceedings of the 30th Brazilian Symposium on Multimedia and the Web (Juiz de Fora/MG). SBC, Porto Alegre, RS, Brasil, 231–239. DOI: 10.5753/webmedia.2024.241857