AI-Powered Risk Control Framework for Health Insurance Fund Management: a Comprehensive Data-Driven Approach

In the revised version, we addressed the reviewers' comments by clarifying data normalization, providing details on data dimensions and sample sizes, and including illustrative examples. The abstract has been restructured for clarity, and the introduction now presents relevant statistics on health insurance fraud, outlines the research gap, and specifies study objectives. We updated the literature review with recent publications and a critical analysis of existing methods. Additionally, we enhanced the methodology section with explanations for algorithm choices and ethical considerations. The results now include comparative performance metrics, a detailed analysis of false positives and negatives, and sensitivity analyses. Finally, the conclusion explicitly states how research objectives were met and offers directions for future work.