XAI-based cross-ensemble feature ranking methodology for machine learning models

Abstract Artificial Intelligence (AI) as one robust technology has been used in various fields, making innovative society possible and changing our lifestyles. However, the black box problem is still big for artificial intelligence. In this study, we first compared results of kernel Shapley Additive exPlanations (SHAP) machine learning models found that single SHAP model cannot explain at human knowledge level. Then factors’ global ranking was calculated using proposed ensemble methodology. Finally, new with other factor method. Our experimental declare cross-ensemble feature methodology provides stable comparatively reliable both classification regression models.