The Optimal Tradeoff between Data Privacy and Utility

under the mentorship of Dr. Wanrong Zhang (postdoctoral fellow at Harvard John A. Paulson School Of Engineering)

Abstract

This research paper investigates the optimal tradeoff between data privacy and utility in the realm of differential privacy, focusing on its application to medical records. Employing advanced privacy-preserving techniques, including pseudonymization, generalization, perturbation, and the strategic introduction of Laplace noise, the study aims to effectively anonymize medical datasets while preserving their analytical utility.

To conduct a comparative analysis of each of these privacy-preserving methods, privacy attacks are simulated on data protected using different privacy-preserving techniques, providing a comprehensive evaluation of the accuracy and robustness of each implemented privacy technique.

Further, this paper also conducts a  legal analysis of pertinent government regulations like the General Data Protection Regulation and the Health Insurance Portability and Accountability Act. Through derived insights from their analysis, the paper makes policy recommendations to strengthen privacy protection in the context of achieving the optimal data privacy and utility tradeoff.