An Improved K-means Clustering Model Based on Support Vector Machine for Health Insurance Cost Prediction

Health insurance is the total amount of medical expenses incurred by the insured's illness. The control of health insurance costs is an important means for managing health insurance expenses. Controlling the cost of medical insurance has become an important issue and problem in the world today, especially in China. Therefore, conducting research on health insurance cost control has become a hot topic in academia and industry, attracting the attention of many scholars. This paper proposes a support vector machine model based on K-means and uses the model to analyse and predict residential health insurance cost data. We use five model evaluation metrics, ME, RMSE, MAE, MPE and MAPE, to verify that the model is effective. From the result, this paper will make some recommendations for the future development of health insurance cost control to further promote health insurance.