This video presented by Achilles Natsis, a Health Research Actuary at the Society of Actuaries Research Institute, who discusses a report on using interpretable machine learning methods to detect health insurance fraud. The report, developed in collaboration with the Sri Sathya Sai Institute of Higher Learning in India, provides a comprehensive framework for applying machine learning to identify fraudulent health insurance claims. The research highlights the significant financial impact of insurance fraud and the potential benefits of using machine learning models to improve fraud detection and reduce costs. The methodology involves cleaning raw data, creating trigger functions, and applying various machine learning models and interpretability techniques to enhance fraud detection. The ultimate goal is to provide health insurance payers with better tools to detect fraud and reduce unnecessary medical costs.
Link to full report : https://www.soa.org/resources/research-reports/2024/interpretable-ml-methods/
Contributors: Achilles Natsis, FSA, MAAA; Rose Northon; Eric Milner, ASA, MAAA; Jon Forster, ASA, MAAA