With health insurance, the financial risk of unexpected medical costs is distributed among many people, reducing the amount of money at risk. Global public health expenditure has almost quadrupled over the last 20 years, and in 2023 it is expected to reach $8.5 trillion, or 9.8% of the world GDP after accounting for inflation. Multinational multi-private sectors supply 70% of outpatient care and 60% of complete medical treatments, sometimes at excessive costs. Health insurance has become a necessary good due to rising healthcare costs, higher life expectancies, and the rise of non-communicable illnesses. The expansion of insurance data availability has made it possible for insurance companies to use predictive modelling to improve their customer service and corporate operations. In order to forecast future output values based on consumer behaviour patterns, insurance policies, data-driven decision-making, and the creation of new schemes, historical insurance data is examined using computer algorithms and machine learning (ML). The insurance business has shown great promise for machine learning (ML), which led to the creation of the ML Health Insurance Prediction System. This cost-price prediction system makes it easier to determine premium values quickly and efficiently, which lowers medical costs. Three regression models are compared and contrasted in this system: Random Forest Regressor, Support Vector Regression, and Linear Regression. The models were trained on a dataset, which allowed for the generation of predictions and the validation of the model's efficacy by comparison with real data