Uncertainty-aware multi-criteria decision analysis for evaluation of explainable artificial intelligence methods: A use case from the healthcare domain

This study introduces a Z-numbers-based Weighted Sum Model (WSM) tailored to evaluate user satisfaction with explanations provided by Explainable Artificial Intelligence (XAI) methods in the healthcare domain. Focusing on the interpretability of XAI, we measure how users perceive the adequacy of explanations through the lens of SHapley Additive exPlanations (SHAP), Individual Conditional Expectation (ICE) plots, and Counterfactual Explanations (CFE). By conducting interviews with healthcare professionals, we integrate their qualitative feedback with quantitative analysis to assess the effectiveness of these methods. The results present a user-centric perspective on the clarity, relevance, and trustworthiness of the generated post-hoc explanations. This study advances the fields of information sciences and healthcare by offering a systematic approach for evaluating XAI, enhancing the transparency and reliability of AI in critical decision-making processes.