A Hybrid Association Rule and Ensemble Unsupervised Learning Framework for Generalizable Healthcare Fraud Detection on South Korea's NHIS

Background: Healthcare insurance fraud continues to drain billions from national health systems each year, making reliable and adaptable detection approaches essential. Many existing models, however, rely heavily on specific datasets such as U.S. Medicare claims, which limits how well they transfer to other healthcare environments with different data structures.Methodology: To address this gap, this study proposes a two-stage unsupervised framework designed for use across diverse health systems. In the first stage, the Apriori algorithm is applied to uncover association rules that describe patterns among patients, providers, and medical services. These interpretable patterns are then assessed in the second stage using an ensemble of four unsupervised anomaly detection models: Isolation Forest, Cluster-Based Local Outlier Factor (CBLOF), Empirical Cumulative Distribution–based Outlier Detection (ECOD), and One-Class SVM. The framework’s flexibility and robustness were tested using the extensive National Health Insurance Service (NHIS) dataset from South Korea, which covers roughly 97% of the country’s population under a single universal insurance system. Results: When applied to NHIS data, the framework successfully identified unusual and potentially fraudulent claim behaviors. Among the models, CBLOF achieved the highest silhouette score (0.118), followed by Isolation Forest (0.101), suggesting strong and consistent performance.Conclusion: The findings indicate that combining association rule mining with unsupervised learning offers a practical and generalizable solution for healthcare fraud detection. Its validation on the NHIS dataset demonstrates adaptability across different insurance models and healthcare structures worldwide.