Explainable Artificial Intelligence Framework for Trustworthy Decision Support Systems

Abstract

The application of artificial intelligence in decision support systems faces transparency challenges that can reduce user confidence in the results of recommendations. This research aims to develop an Explainable Artificial Intelligence framework that supports a transparent, understandable, and accountable decision support system. The research used an exploratory qualitative approach with semi-structured interviews with 20 informants consisting of AI developers, information systems experts, and professional users. The data was analyzed using thematic analysis to identify the need for explanation, confidence-building factors, and the characteristics of information that were considered relevant. The results show that clarity of decision reasons, consistency of explanations, traceability of processes, and suitability of context are the main elements of forming trust. This research produces an XAI conceptual framework that can be the basis for the development of a more transparent, reliable, and user-oriented decision support system