Insurance Pricing from the Perspective of Healthcare Cost Prediction—— An Empirical Analysis Based on Machine Learning Using UCI Data

DOI

10.2991/978-94-6239-801-6_132How to use a DOI?

Keywords

Health Insurance Pricing; Medical Cost Prediction; Machine Learning; Xgboost; Algorithmic Fairness

Abstract

In the process of continuous development of the medical insurance industry, the pricing mechanism of medical insurance is gradually maturing. In this context, it has become a key issue to carry out more scientific and reasonable medical cost prediction and insurance pricing based on individual risk characteristics. Based on the UCI medical insurance data set, this study carefully selected age, gender BMI、 Smoking status, number of children, region and other representative variables. Using linear regression, random forest and XGBoost, three common and effective model building methods, this paper makes an in-depth study of individual annual medical expenses. By comparing and analyzing the results of different models, this paper comprehensively analyzes the influencing factors of individual annual medical expenses and the prediction effect of each model. Finally, it was found that smoking, age and BMI were the core factors affecting medical expenses, and smoking had the most prominent impact on medical expenses. The effect of BMI on medical expenses is nonlinear, and the ability of the number of children to explain the change of medical expenses is relatively weak.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zetao Shi
PY  - 2026
DA  - 2026/10/23
TI  - Insurance Pricing from the Perspective of Healthcare Cost Prediction—— An Empirical Analysis Based on Machine Learning Using UCI Data
BT  - Proceedings of the 2026 2nd International Conference on Financial Innovation and Marketing Management (FIMM 2026)
PB  - Atlantis Press
SP  - 1271
EP  - 1280
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-801-6_132
DO  - 10.2991/978-94-6239-801-6_132
ID  - Shi2026
ER  -