O Paradoxo da Hibridização: O Impacto da Combinação de Algoritmos na Justiça e no Risco em Sistemas de Recomendação
Resumo
Este trabalho investiga o impacto da hibridização ponderada em Sistemas de Recomendação sob as óticas de justiça (fairness) e sensibilidade ao risco. Empregando uma avaliação rigorosa com 20 janelas temporais deslizantes na base MovieLens, analisou-se como a combinação de modelos via regressão afeta a equidade e a robustez das sugestões. Os resultados evidenciam um fenômeno contraintuitivo: a hibridização atuou como um nivelador genérico, intensificando as disparidades de desempenho entre grupos e elevando o risco global do sistema. Tais achados demonstram que a otimização focada na minimização do erro médio sacrifica a personalização fina, prejudicando usuários com perfis complexos ou pertencentes a grupos minoritários.
Palavras-chave:
Sistemas de Recomendação, Hibridização Ponderada, Justiça Algorítmica, Sensibilidade ao Risco, Avaliação Temporal
Referências
Angwin, J., Larson, J., Mattu, S., and Kirchner, L. (2016). How we analyzed the COMPAS recidivism algorithm. ProPublica.
Bao, X., Bergman, L., and Thompson, R. (2009). Stacking Recommendation Engines with Additional Meta-features. In ACM RecSys, pages 109–116, New York, NY, USA.
Boratto, L. et al. (2025). Popularity bias in recommender systems: The search for fairness in the long tail. Information, 16(2).
Burke, R. (2002). Hybrid recommender systems: Survey and experiments. User Modeling and User-Adapted Interaction, 12(4):331–370.
Burke, R., Sonboli, N., and Ordonez-Gauger, A. (2018). Balanced neighborhoods for multi-sided fairness in recommendation. In Friedler, S. A. and Wilson, C., editors, Proceedings of the 1st Conference on Fairness, Accountability and Transparency, volume 81 of Proceedings of Machine Learning Research, pages 202–214. PMLR.
Deldjoo, Y., Jannach, D., Bellogin, A., Difonzo, A., and Zanzonelli, D. (2023). Fairness in recommender systems: Research landscape and future directions. User Modeling and User-Adapted Interaction.
Dinçer, B. T., Ounis, I., and Macdonald, C. (2014). Tackling biased baselines in the risk-sensitive evaluation of retrieval systems. In Proceedings of the 36th European Conference on Information Retrieval, pages 26–38. Springer.
Farnadi, G., Kouki, P., Thompson, S. K., Srinivasan, S., and Getoor, L. (2018). A fairness-aware hybrid recommender system. CoRR, abs/1809.09030.
Fortes, R. S. (2022). Enhancing the multi-objective recommendation from three new perspectives: data characterization, risk-sensitiveness, and prioritization of the objectives.
Fu, Z., Xian, Y., Gao, R., Zhao, J., Huang, Q., Ge, Y., Xu, S., Geng, S., Shah, C., Zhang, Y., and de Melo, G. (2020). Fairness-aware explainable recommendation over knowledge graphs.
Herlocker, J., Konstan, J. A., Borchers, A., and Riedl, J. (2001). Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems (TOIS), 19(1):5–53.
Ji, Y., Sun, A., Zhang, J., and Li, C. (2023). A critical study on data leakage in recommender system offline evaluation. ACM Transactions on Information Systems (TOIS), 41(3).
Klimashevskaia, A., Jannach, D., Elahi, M., and Trattner, C. (2024). A survey on popularity bias in recommender systems. User Modeling and User-Adapted Interaction.
Ma, H., Zhou, D., Liu, C., Lyu, M. R., and King, I. (2011). Recommender systems with social regularization. In Proceedings of the Fourth ACM International Conference on Web Search and Data Mining, WSDM ’11, pages 287–296, New York, NY, USA. Association for Computing Machinery.
Pitoura, E., Stefanidis, K., and Koutrika, G. (2021). Fairness in rankings and recommendations: an overview. The VLDB Journal, pages 651–654.
Quadrana, M., Cremonesi, P., and Jannach, D. (2018). Sequence-aware recommender systems. ACM Computing Surveys (CSUR), 51(4):1–36.
Sill, J., Takacs, G., Mackey, L., and Lin, D. (2009). Feature-Weighted Linear Stacking. arXiv:0911.0460 [cs].
Wang, L., Bennett, P. N., and Collins-Thompson, K. (2012). Robust ranking models via risk-sensitive optimization.
Wang, Y., Chen, J., et al. (2023). A survey on fairness-aware recommender systems. Information Fusion.
Bao, X., Bergman, L., and Thompson, R. (2009). Stacking Recommendation Engines with Additional Meta-features. In ACM RecSys, pages 109–116, New York, NY, USA.
Boratto, L. et al. (2025). Popularity bias in recommender systems: The search for fairness in the long tail. Information, 16(2).
Burke, R. (2002). Hybrid recommender systems: Survey and experiments. User Modeling and User-Adapted Interaction, 12(4):331–370.
Burke, R., Sonboli, N., and Ordonez-Gauger, A. (2018). Balanced neighborhoods for multi-sided fairness in recommendation. In Friedler, S. A. and Wilson, C., editors, Proceedings of the 1st Conference on Fairness, Accountability and Transparency, volume 81 of Proceedings of Machine Learning Research, pages 202–214. PMLR.
Deldjoo, Y., Jannach, D., Bellogin, A., Difonzo, A., and Zanzonelli, D. (2023). Fairness in recommender systems: Research landscape and future directions. User Modeling and User-Adapted Interaction.
Dinçer, B. T., Ounis, I., and Macdonald, C. (2014). Tackling biased baselines in the risk-sensitive evaluation of retrieval systems. In Proceedings of the 36th European Conference on Information Retrieval, pages 26–38. Springer.
Farnadi, G., Kouki, P., Thompson, S. K., Srinivasan, S., and Getoor, L. (2018). A fairness-aware hybrid recommender system. CoRR, abs/1809.09030.
Fortes, R. S. (2022). Enhancing the multi-objective recommendation from three new perspectives: data characterization, risk-sensitiveness, and prioritization of the objectives.
Fu, Z., Xian, Y., Gao, R., Zhao, J., Huang, Q., Ge, Y., Xu, S., Geng, S., Shah, C., Zhang, Y., and de Melo, G. (2020). Fairness-aware explainable recommendation over knowledge graphs.
Herlocker, J., Konstan, J. A., Borchers, A., and Riedl, J. (2001). Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems (TOIS), 19(1):5–53.
Ji, Y., Sun, A., Zhang, J., and Li, C. (2023). A critical study on data leakage in recommender system offline evaluation. ACM Transactions on Information Systems (TOIS), 41(3).
Klimashevskaia, A., Jannach, D., Elahi, M., and Trattner, C. (2024). A survey on popularity bias in recommender systems. User Modeling and User-Adapted Interaction.
Ma, H., Zhou, D., Liu, C., Lyu, M. R., and King, I. (2011). Recommender systems with social regularization. In Proceedings of the Fourth ACM International Conference on Web Search and Data Mining, WSDM ’11, pages 287–296, New York, NY, USA. Association for Computing Machinery.
Pitoura, E., Stefanidis, K., and Koutrika, G. (2021). Fairness in rankings and recommendations: an overview. The VLDB Journal, pages 651–654.
Quadrana, M., Cremonesi, P., and Jannach, D. (2018). Sequence-aware recommender systems. ACM Computing Surveys (CSUR), 51(4):1–36.
Sill, J., Takacs, G., Mackey, L., and Lin, D. (2009). Feature-Weighted Linear Stacking. arXiv:0911.0460 [cs].
Wang, L., Bennett, P. N., and Collins-Thompson, K. (2012). Robust ranking models via risk-sensitive optimization.
Wang, Y., Chen, J., et al. (2023). A survey on fairness-aware recommender systems. Information Fusion.
Publicado
08/09/2026
Como Citar
COSTA, João Paulo Prata; FERREIRA, Daniel José Chaves; FERREIRA, Anderson A.; FORTES, Reinaldo Silva.
O Paradoxo da Hibridização: O Impacto da Combinação de Algoritmos na Justiça e no Risco em Sistemas de Recomendação. In: SIMPÓSIO BRASILEIRO DE BANCO DE DADOS (SBBD), 41. , 2026, São Carlos/SP.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 414-426.
ISSN 2763-8979.
DOI: https://doi.org/10.5753/sbbd.2026.249230.
