Predicting health insurance premiums using machine learning: A novel regression-based model for enhanced accuracy and personalization

Machine learning (ML) is reshaping healthcare insurance by streamlining the prediction of health insurance premiums, allowing insurers to offer more personalized and efficient services to consumers. This paper presents a novel regression-based model designed to accurately forecast health insurance costs based on individual characteristics, bridging the gap between insurers and policyholders. Leveraging an artificial neural network (ANN), the model considers key factors, including age, gender, body mass index, number of dependents, smoking status, and geographic location, to predict premium costs with greater precision. Our approach demonstrates an advancement over traditional methods, achieving a prediction accuracy of 92.72% in experimental trials. This high performance underscores the model’s capability to provide tailored premium estimations, thus enhancing customer satisfaction by offering fair and data-driven pricing. This research further evaluates the model’s efficacy through key performance metrics, confirming its robustness and practical applicability for insurers aiming to adopt ML for personalized healthcare coverage. The proposed model contributes to the field of digital health insurance, offering a scalable and data-rich approach to premium estimation that benefits both insurers and consumers in today’s tech-driven healthcare landscape.