Dear authors, Thanks for your submission. One of the associate editors has read the paper and we have decided not to send this out for review. While there are some potentially impactful ideas in this paper, ultimately we did not think it would be accepted by reviewers. Some high-level comments that we believe would improve the paper: 1) Make it clearer what this paper achieves beyond existing work in fair multi-class classification papers, such as [1-4] below 2) The experiments show some interesting results, for an article in this area to be published in AIJ, reviewers would likely expect to see some theoretical results that give e.g. bounds on the performance and/or fairness of the resulting classifier; see e.g. Alghamdi et al [2]. 3) Regarding the experimental design, I would suggest at least comparing with Alghamdi et al [2], but also, in the absence of multi-class classifiers, they include some additional baselines by tweaking existing methods. 4) Why not use HSLS and ENEM datasets used by Alghamdi et al [2] as well? We hope that these comments can be used to improve the work [1] Zhao, Shengjia, et al. "Calibrating predictions to decisions: A novel approach to multi-class calibration." Advances in Neural Information Processing Systems 34 (2021): 22313-22324. [2] Alghamdi, Wael, et al. "Beyond adult and compas: Fairness in multi-class prediction." 36th Conference on Neural Information Processing Systems (NeurIPS 2022) [3] Mishra, Pradeepta. "AI Model Fairness Using a What-If Scenario." Practical Explainable AI Using Python: Artificial Intelligence Model Explanations Using Python-based Libraries, Extensions, and Frameworks. Berkeley, CA: Apress, 2021. 229-242. [4] Rouzot, Julien, Julien Ferry, and Marie-José Huguet. "Learning Optimal Fair Scoring Systems for Multi-Class Classification." 2022 IEEE 34th International Conference on Tools with Artificial Intelligence (ICTAI). IEEE, 2022.
25-09-2023 Обдумать, что можно добавить в статью, чтобы ее можно было отправить в журнал Q1.Григорий Ясновидов 25-09-2023 Разбить на подзадачи с исполнителями и обдумать примерные сроки.Григорий Ясновидов 25-09-2023 Подать статью в журнал Q1.Григорий Ясновидов 25-09-2023 — 23-10-2023 Учесть комментарии и переподать в другой журнал Q1Григорий Ясновидов
Dear authors, Thanks for your submission. One of the associate editors has read the paper and we have decided not to send this out for review. While there are some potentially impactful ideas in this paper, ultimately we did not think it would be accepted by reviewers. Some high-level comments that we believe would improve the paper: 1) Make it clearer what this paper achieves beyond existing work in fair multi-class classification papers, such as [1-4] below 2) The experiments show some interesting results, for an article in this area to be published in AIJ, reviewers would likely expect to see some theoretical results that give e.g. bounds on the performance and/or fairness of the resulting classifier; see e.g. Alghamdi et al [2]. 3) Regarding the experimental design, I would suggest at least comparing with Alghamdi et al [2], but also, in the absence of multi-class classifiers, they include some additional baselines by tweaking existing methods. 4) Why not use HSLS and ENEM datasets used by Alghamdi et al [2] as well? We hope that these comments can be used to improve the work [1] Zhao, Shengjia, et al. "Calibrating predictions to decisions: A novel approach to multi-class calibration." Advances in Neural Information Processing Systems 34 (2021): 22313-22324. [2] Alghamdi, Wael, et al. "Beyond adult and compas: Fairness in multi-class prediction." 36th Conference on Neural Information Processing Systems (NeurIPS 2022) [3] Mishra, Pradeepta. "AI Model Fairness Using a What-If Scenario." Practical Explainable AI Using Python: Artificial Intelligence Model Explanations Using Python-based Libraries, Extensions, and Frameworks. Berkeley, CA: Apress, 2021. 229-242. [4] Rouzot, Julien, Julien Ferry, and Marie-José Huguet. "Learning Optimal Fair Scoring Systems for Multi-Class Classification." 2022 IEEE 34th International Conference on Tools with Artificial Intelligence (ICTAI). IEEE, 2022.