<![CDATA[The increasing complexity of organizational decision-making, driven by heterogeneous data, evolving user preferences, and dynamic business environments, has exposed the limitations of conventional Decision Support Systems (DSS). Traditional DSS rely on static decision models and predefined preferences, limiting their adaptability and personalization. Although Artificial Intelligence (AI)-based DSS have improved predictive capabilities, many still lack adaptive preference learning, continuous feedback, explainability, and lifelong learning. This study aims to develop a Foundational Framework for Intelligent Data-Driven Decision Support Systems (ID-DSS) based on Adaptive Preference Learning (APL). The research adopts the Design Science Research (DSR) methodology, incorporating a systematic literature review, problem identification, requirement analysis, framework design, and conceptual validation. The proposed framework integrates data analytics, adaptive preference learning, decision intelligence, explainable AI, continuous feedback, and knowledge updating within a closed-loop learning architecture. The Adaptive Preference Learning mechanism continuously refines user preferences using explicit feedback, implicit behavioral observations, contextual information, and incremental learning, enabling recommendations to become increasingly personalized and adaptive. Furthermore, explainable AI enhances transparency by providing interpretable reasoning for recommendation outcomes. The proposed framework establishes a theoretical foundation for next-generation intelligent DSS that are adaptive, personalized, transparent, context-aware, and capable of continuous learning, with potential applications across healthcare, finance, manufacturing, education, smart cities, and public administration.]]>
