<![CDATA[Rapid expansion of Islamic Financial Technology (Islamic FinTech) has increased the complexity of ensuring continuous Sharia compliance across digital financial products, investment services, and automated transactions. Conventional compliance assessment primarily depends on manual evaluation by Sharia scholars and supervisory boards, creating challenges related to scalability, consistency, operational efficiency, and timely risk identification. This study aimed to develop and evaluate a machine learning framework for automated Sharia compliance assessment and financial risk screening that integrates predictive intelligence with explainable and transparent decision support. A mixed-methods sequential explanatory research design was employed, combining quantitative analysis of 125,000 anonymized financial transaction records using supervised machine learning algorithms with qualitative evidence obtained from expert interviews, institutional document analysis, and regulatory validation. Comparative evaluation demonstrated that the proposed framework achieved high predictive performance, with XGBoost providing the highest classification accuracy while Explainable Artificial Intelligence techniques enhanced transparency through interpretable decision explanations. Qualitative findings confirmed that automated screening significantly reduced compliance review time, strengthened institutional consistency, and improved stakeholder confidence without replacing the essential role of Sharia scholars in complex jurisprudential decisions. Results indicate that algorithmic Sharia compliance functions most effectively as a human-centered decision-support framework integrating machine learning, Islamic jurisprudence, financial governance, and explainable artificial intelligence. Responsible implementation of this framework provides a scalable pathway toward trustworthy Islamic FinTech, enhanced regulatory accountability, and sustainable digital financial innovation.]]>
Copyrights © 2026
