o:35558 OPPORTUNITIES FOR HEALTHCARE COST PREDICTION USING MACHINE LEARNING ALGORITHMS en Abstract: The growing trend of healthcare costs, increased life expectancy, and the increasing availability of data on policyholders indicate the importance of the application of machine learning in health insurance. Using historical data of policyholders, machine learning enables the prediction of healthcare costs, identification of high-risk individuals for hospitalisation, assessment of the likelihood of chronic diseases, and more. The subject of research in this paper are the opportunities for healthcare cost prediction by implementing different machine learning algorithms. Based on the public database from the Kaggle website, the created model incorporates various machine learning algorithms such as Random Forest, Gradient Boosting and Linear Regression. The aim of the paper is to point out that selecting a predictive machine learning model with the best performance can significantly improve the prediction of individual healthcare costs. This, in turn, contributes to determining appropriate premiums for voluntary health insurance. Keywords: healthcare costs, health insurance, insurance premium, machine learning, database, Random Forest, Gradient Boosting, Linear Regression. 1552099 10.24867/SYMOPIS-2024-51-096 2025-01-14T10:19:57.387Z 44 no 46 Tatjana Rakonjac-Antić Univerzitet u Beogradu Ekonomski fakultet 0000-0003-0371-0115 46 Marija Koprivica Univerzitet u Beogradu Ekonomski fakultet 0000-0003-4239-2252 46 Milica Kočović De Santo Institute of Economics Sciences, Belgrade 0000-0003-3304-7801 46 Kristina Bradić application/pdf 1431630 https://phaidrabg.bg.ac.rs/o:35558 no yes 18 70 92000004 11A03 51. Симпозијум о операционим истраживањима 51 602 607 2024