EXPLORING THE ROLE OF SYNTHETIC DATA GENERATION IN TRAINING ROBUST INSURANCE MODELS AND MITIGATING DATA PRIVACY CONCERNS

Authors

  • Aguna Triayudi English Author

Keywords:

information, efficacy, Adversarial

Abstract

The insurance industry's growing dependence on data-driven models highlights the need 
for practical ways to resolve data privacy issues and improve model resilience. With an 
emphasis on techniques like Generative Adversarial Networks (GANs) and Variational 
Autoencoders (VAEs), this study explores the use of synthetic data creation in training 
insurance models. Synthetic data delivers equivalent efficacy while resolving privacy 
concerns, according to the research, which compares the performance of models trained 
with synthetic data to those taught with actual data. Synthetic data reduces the risks 
involved with managing sensitive information by using strategies including 
anonymization, de-identification, and differential privacy. The findings imply that 
synthetic data might be used as a useful instrument to improve model accuracy and data 
privacy in the insurance industry. The results demonstrate how synthetic data may be 
used to strike a compromise between privacy and data value, encouraging safer and more 
effective data management techniques.

Downloads

Published

2025-06-30