Beyond Algorithmic Disagreement: Stabilizing Counterfactual Explanations in Tree-Based Models. A Case Study
Resumo
A Inteligência Artificial tem revolucionado o suporte à decisão clínica em doenças tropicais negligenciadas como a Leishmaniose Visceral (LV). Contudo, a opacidade desses modelos e a geração de contrafactuais clinicamente inviáveis dificultam sua adoção prática. Este trabalho propõe um pipeline de explicabilidade acionável baseado no framework DiCE integrado a um Gradient Boosting Machine (GBM), aplicando restrições causais, suavização probabilística e custos assimétricos. Os resultados indicam que a suavização estabilizou as recomendações, atingindo 96,81% de similaridade de trajetória e reduzindo o problema do desacordo em 44,44%. Por outro lado, a imposição de custos gerou um deslocamento residual de apenas -0,27% nas condutas de alto risco, consolidando os biomarcadores laboratoriais como verdadeiras âncoras matemáticas mandatórias para a reversão do desfecho crítico. O modelo alcançou uma taxa de validade coletiva de 83,33% nas soluções.Referências
Artelt, A., Vaquet, V., Velioglu, R., Hinder, F., Brinkrolf, J., Schilling, M., and Hammer, B. (2021). Evaluating robustness of counterfactual explanations. In Proceedings of the 2021 IEEE Symp Series on Comp Intelligence, pages 1–8. IEEE.
Batista, F. M. d. A., Sousa, R. A. d., Mendonça, V. J., Aguiar, B. G. A., Ibiapina, A. B., de Araújo Albuquerque, L. P., and Costa, C. H. N. (2021). Perfil epidemiológico e tendência temporal da leishmaniose visceral: Piauí, brasil, 2008 a 2018. Carderno em Saúde Pública, 37(11).
Brughmans, D., Melis, L., and Martens, D. (2024). Disagreement amongst counterfactual explanations: how transparency can be misleading. TOP, 32(2):429–462.
Carrizosa, E., Ramírez-Ayerbe, J., and Romero Morales, D. (2024). Generating collective counterfactual explanations in score-based classification via mathematical optimization. Expert Systems with Applications, 238:121954.
Costa, D. L., Rocha, R. L., Chaves, E. d. B. F., Batista, V. G. d. V., Costa, H. L., and Costa, C. H. N. (2016). Predicting death from kala-azar: construction, development, and validation of a score set and accompanying software. Revista da Sociedade Brasileira de Medicina Tropical, 49(6):728–740.
Del Ser, J., Barredo-Arrieta, A., Díaz-Rodríguez, N., Herrera, F., Sarantis, A., and Holzinger, A. (2024). On generating trustworthy counterfactual explanations. Information Sciences, 655:119898.
Guo, Y., Hao, Z., Zhao, S., Gong, J., and Yang, F. (2020). Artificial intelligence in health care: Bibliometric analysis. Journal of Medical Internet Research, 22(7):e18228.
Hudec, M., Minarikova, E., Mesiar, R., et al. (2021). Classification by ordinal sums of conjunctive and disjunctive functions for explainable AI and interpretable machine learning solutions. Knowledge-Based Systems, 220:106916.
Jiang, J., Leofante, F., Rago, A., et al. (2024). Robust counterfactual explanations in machine learning: a survey. arXiv preprint arXiv:2402.01928.
Karimi, A.-H., Barthe, G., Schölkopf, B., and Valera, I. (2021). Algorithmic recourse: From counterfactual explanations to interventions. ACM Computing Surveys (CSUR), 53(6):1–17.
Lee, C. S. and Lee, A. Y. (2020). Clinical applications of continual learning machine learning. The Lancet Digital Health, 2(6):e279–e281.
Lee, M. H. and Chew, C. J. (2023). Understanding the effect of counterfactual explanations on trust and reliance on ai for human-ai collaborative clinical decision making. In Proceedings of the 2023 ACM Designing Interactive Systems Conference, pages 1–15.
Lee, S., Mohr, N. M., Street, W. N., and Nadkarni, P. (2019). Machine learning in relation to emergency medicine clinical and operational scenarios: An overview. Western Journal of Emergency Medicine, 20(2):219–227.
Lenatti, M., Carlevaro, A., Guergachi, A., Keshavjee, K., Mongelli, M., and Paglialonga, A. (2025). Estimation and conformity evaluation of multi-class counterfactual explanations for chronic disease prevention. IEEE Journal of Biomedical and Health Informatics, 29(9):6132–6142.
Mothilal, R. K., Sharma, A., and Tan, C. (2020). Explaining machine learning classifiers through diverse counterfactual explanations. In Proceedings of the 2020 Conf. on Fairness, Accountability, and Transparency, pages 607–617.
Nohara, Y., Matsumoto, K., Soejima, H., and Nakashima, N. (2022). Explanation of machine learning models using shapley additive explanation and application for real data in hospital. Computer Methods and Programs in Biomedicine, 214:106584.
Pfeifer, B., Krzyzinski, M., Baniecki, H., Holzinger, A., and Biecek, P. (2025). Explaining and visualizing black-box models through counterfactual paths. Pattern Analysis and Applications, 28(154):1–19.
Rasouli, P. and Yu, I. C. (2021). Analyzing and improving the robustness of tabular classifiers using counterfactual explanations. In Proceedings of the 2021 20th IEEE International Conf on ML and Applications, pages 1–8. IEEE.
Saranti, A., Hudec, M., Minarikova, E., et al. (2022). Actionable explainable AI (AXAI): A practical example with aggregation functions for adaptive classification and textual explanations for interpretable machine learning. Machine Learning and Knowledge Extraction, 4(4):924–953.
Ustun, B., Spangher, A., and Liu, Y. (2019). Actionable recourse in linear classification. In Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT*), pages 10–19. ACM.
Verma, S., Boonsanong, V., Hoang, M., Hines, K., Dickerson, J., and Shah, C. (2024). Counterfactual explanations and algorithmic recourses for machine learning: A review. ACM Computing Surveys, 56(312):1–42.
Wachter, S., Mittelstadt, B., and Russell, C. (2017). Counterfactual explanations without opening the black box: Automated decisions and the GDPR. Harvard Journal of Law & Technology, 31:841.
Batista, F. M. d. A., Sousa, R. A. d., Mendonça, V. J., Aguiar, B. G. A., Ibiapina, A. B., de Araújo Albuquerque, L. P., and Costa, C. H. N. (2021). Perfil epidemiológico e tendência temporal da leishmaniose visceral: Piauí, brasil, 2008 a 2018. Carderno em Saúde Pública, 37(11).
Brughmans, D., Melis, L., and Martens, D. (2024). Disagreement amongst counterfactual explanations: how transparency can be misleading. TOP, 32(2):429–462.
Carrizosa, E., Ramírez-Ayerbe, J., and Romero Morales, D. (2024). Generating collective counterfactual explanations in score-based classification via mathematical optimization. Expert Systems with Applications, 238:121954.
Costa, D. L., Rocha, R. L., Chaves, E. d. B. F., Batista, V. G. d. V., Costa, H. L., and Costa, C. H. N. (2016). Predicting death from kala-azar: construction, development, and validation of a score set and accompanying software. Revista da Sociedade Brasileira de Medicina Tropical, 49(6):728–740.
Del Ser, J., Barredo-Arrieta, A., Díaz-Rodríguez, N., Herrera, F., Sarantis, A., and Holzinger, A. (2024). On generating trustworthy counterfactual explanations. Information Sciences, 655:119898.
Guo, Y., Hao, Z., Zhao, S., Gong, J., and Yang, F. (2020). Artificial intelligence in health care: Bibliometric analysis. Journal of Medical Internet Research, 22(7):e18228.
Hudec, M., Minarikova, E., Mesiar, R., et al. (2021). Classification by ordinal sums of conjunctive and disjunctive functions for explainable AI and interpretable machine learning solutions. Knowledge-Based Systems, 220:106916.
Jiang, J., Leofante, F., Rago, A., et al. (2024). Robust counterfactual explanations in machine learning: a survey. arXiv preprint arXiv:2402.01928.
Karimi, A.-H., Barthe, G., Schölkopf, B., and Valera, I. (2021). Algorithmic recourse: From counterfactual explanations to interventions. ACM Computing Surveys (CSUR), 53(6):1–17.
Lee, C. S. and Lee, A. Y. (2020). Clinical applications of continual learning machine learning. The Lancet Digital Health, 2(6):e279–e281.
Lee, M. H. and Chew, C. J. (2023). Understanding the effect of counterfactual explanations on trust and reliance on ai for human-ai collaborative clinical decision making. In Proceedings of the 2023 ACM Designing Interactive Systems Conference, pages 1–15.
Lee, S., Mohr, N. M., Street, W. N., and Nadkarni, P. (2019). Machine learning in relation to emergency medicine clinical and operational scenarios: An overview. Western Journal of Emergency Medicine, 20(2):219–227.
Lenatti, M., Carlevaro, A., Guergachi, A., Keshavjee, K., Mongelli, M., and Paglialonga, A. (2025). Estimation and conformity evaluation of multi-class counterfactual explanations for chronic disease prevention. IEEE Journal of Biomedical and Health Informatics, 29(9):6132–6142.
Mothilal, R. K., Sharma, A., and Tan, C. (2020). Explaining machine learning classifiers through diverse counterfactual explanations. In Proceedings of the 2020 Conf. on Fairness, Accountability, and Transparency, pages 607–617.
Nohara, Y., Matsumoto, K., Soejima, H., and Nakashima, N. (2022). Explanation of machine learning models using shapley additive explanation and application for real data in hospital. Computer Methods and Programs in Biomedicine, 214:106584.
Pfeifer, B., Krzyzinski, M., Baniecki, H., Holzinger, A., and Biecek, P. (2025). Explaining and visualizing black-box models through counterfactual paths. Pattern Analysis and Applications, 28(154):1–19.
Rasouli, P. and Yu, I. C. (2021). Analyzing and improving the robustness of tabular classifiers using counterfactual explanations. In Proceedings of the 2021 20th IEEE International Conf on ML and Applications, pages 1–8. IEEE.
Saranti, A., Hudec, M., Minarikova, E., et al. (2022). Actionable explainable AI (AXAI): A practical example with aggregation functions for adaptive classification and textual explanations for interpretable machine learning. Machine Learning and Knowledge Extraction, 4(4):924–953.
Ustun, B., Spangher, A., and Liu, Y. (2019). Actionable recourse in linear classification. In Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT*), pages 10–19. ACM.
Verma, S., Boonsanong, V., Hoang, M., Hines, K., Dickerson, J., and Shah, C. (2024). Counterfactual explanations and algorithmic recourses for machine learning: A review. ACM Computing Surveys, 56(312):1–42.
Wachter, S., Mittelstadt, B., and Russell, C. (2017). Counterfactual explanations without opening the black box: Automated decisions and the GDPR. Harvard Journal of Law & Technology, 31:841.
Publicado
19/10/2026
Como Citar
MARCIANO, Juan M. V.; MACHADO, Vinicius P.; COSTA, Dorcas L.; MORAIS, Ygor F. C.; AMORIM, Anthony Rios.
Beyond Algorithmic Disagreement: Stabilizing Counterfactual Explanations in Tree-Based Models. A Case Study. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 682-693.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.30095.
