<p>In the discussion by Adoko concerning our paper titled “Refined Approaches for Open Stope Stability Analysis in Mining Environments: Hybrid SVM Model with Multi-optimization Strategies and GP Technique” (Huang and Zhou, 2024b; 57:9781–9804), key concerns are raised regarding the applicability of the dataset, the stability condition labeling scheme, and the model performance. In response, the authors provide a detailed clarification. In terms of data, although the dataset used includes instances from multiple mining methods, this does not imply that the developed stability analysis techniques are ineffective. On the contrary, the diversity of the dataset enhances the modeling potential of machine learning, and the developed predictive platform supports customized evaluations based on on-site data. Regarding the numbering scheme for stability conditions, the authors have demonstrated that the numbering convention used in the original study is reasonable, as it aligns with common practices in binary classification and enhances interpretability. Regarding model performance, we have clarified that the newly developed model exhibits robust generalization capabilities and has been validated using the dataset compiled by (Adoke et al. 2022; 40:677–696). Furthermore, the analytical framework proposed by the authors, which combines learning modeling with function mining, provides a “dual insurance” mechanism for the reliability of the output. In conclusion, this paper reaffirms the validity of the proposed stope stability analysis method and emphasizes the need for caution in the future use of the extended stability graph database in related research. This discussion and response serve to clarify potential ambiguities within the original study and provide potential guidance for future research directions.</p>