The Security of Medical Data on Internet Based on Differential Privacy Technology
Zhihan Lv, Francesco Piccialli
Abstract
The study aims at discussing the security of medical data in the Internet era. By using k-anonymity (K-A) and differential privacy (DP), an algorithm model combining K-A and DP was proposed, which was simulated through the experiments. In the Magic and EIA datasets, the algorithm constructed was compared with K-A and the L-diversity model to verify the performance of the model. The model constructed based on DP had the lowest privacy-leakage risks, which increased with the number of identifiers in the Magic and EIA datasets, and the information disclosure was the least. In addition, in its usability analysis, it was found that its value was the most obviously improved and its operation efficiency was the highest. The K-A-DP algorithm can effectively reduce the risk of privacy leakage and information loss, and has achieved excellent results. Despite the deficiencies in the process of the experiment, the study still provides a reference for solving the problem of medical data security.