Scholarlink

Abstract <p>This paper proposes a hybrid Explainable AI architecture that fuses a fully differentiable neuro-fuzzy GD–ANFIS model with the post-hoc SHAP method. The integration is designed to meet XAI 2.0 principles, which call for explanations that are transparent, verifiable, and adaptable at the same time. GD–ANFIS produces human-readable Takagi–Sugeno rules, ensuring structural interpretability, whereas SHAP delivers quantitative feature contributions that are derived from Shapley theory. To merge these layers, we introduce a comparative-audit mechanism that automatically matches the sets of key features identified by both methods, checks whether the directions of influence coincide, and assesses the consistency between SHAP numerical scores and GD–ANFIS linguistic rules. In regression tests on the Boston Housing dataset and surface-water-quality monitoring, RMSE values of 2.30 and 2.36 were obtained, respectively, all with full interpretability preserved. In every case, top-feature overlap between the two explanation layers exceeded 60%, demonstrating strong agreement between structural and numerical interpretations. The proposed architecture therefore offers a practical foundation for responsible XAI 2.0 deployment in critical domains ranging from medicine and ecology to geoinformation systems and finance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Verified Explainability Core: A GD–ANFIS/SHAP Hybrid Architecture for XAI 2.0

  • 期刊论文

  • 发表日期：

    2025年12月1日

查看全文
  • Y. V. Trofimov,
  • A. D. Lebedev,
  • A. S. Ilin,
  • A. N. Averkin

摘要

Abstract

This paper proposes a hybrid Explainable AI architecture that fuses a fully differentiable neuro-fuzzy GD–ANFIS model with the post-hoc SHAP method. The integration is designed to meet XAI 2.0 principles, which call for explanations that are transparent, verifiable, and adaptable at the same time. GD–ANFIS produces human-readable Takagi–Sugeno rules, ensuring structural interpretability, whereas SHAP delivers quantitative feature contributions that are derived from Shapley theory. To merge these layers, we introduce a comparative-audit mechanism that automatically matches the sets of key features identified by both methods, checks whether the directions of influence coincide, and assesses the consistency between SHAP numerical scores and GD–ANFIS linguistic rules. In regression tests on the Boston Housing dataset and surface-water-quality monitoring, RMSE values of 2.30 and 2.36 were obtained, respectively, all with full interpretability preserved. In every case, top-feature overlap between the two explanation layers exceeded 60%, demonstrating strong agreement between structural and numerical interpretations. The proposed architecture therefore offers a practical foundation for responsible XAI 2.0 deployment in critical domains ranging from medicine and ecology to geoinformation systems and finance.