Title: Data privacy and anonymisation of simulated health-care dataset using the ARX open source tool

Authors: Yogesh Beeharry; Noorsabah Y. Fakeeroodeen; Tulsi Pawan Fowdur

Addresses: Faculty of Engineering, Department of Electrical and Electronic Engineering, University of Mauritius, Réduit 80837, Mauritius ' Faculty of Engineering, Department of Electrical and Electronic Engineering, University of Mauritius, Réduit 80837, Mauritius ' Faculty of Engineering, Department of Electrical and Electronic Engineering, University of Mauritius, Réduit 80837, Mauritius

Abstract: Internet of things (IoT) and big data analytics fields have been adopted in many practical applications geared towards the concept of smart cities and smart world. One critical aspect of these concepts is that of smart-health where private health information has to be transmitted over networks. The main concern arising in this scenario is that of data privacy and anonymity. Various open source tools and algorithms have been developed along this line. In this work, an analysis of the K-anonymity and differential privacy alongside t-closeness for sensitive attributes, algorithms have been performed on a simulated health-care dataset using the ARX open-source tool. The results demonstrate that the two models with t-closeness provided similar output results in relation to risk analysis, utility statistics and output data indicating equal appropriateness for both models. However, their property of transformation differed from each other.

Keywords: data anonymisation; privacy; K-anonymity; differential privacy; t-closeness; ARX open-source tool; healthcare dataset; distribution of risks; attributes identification; Quasi-identifiers.

DOI: 10.1504/IJSSC.2023.133235

International Journal of Space-Based and Situated Computing, 2023 Vol.9 No.3, pp.125 - 137

Received: 07 Apr 2019
Accepted: 11 May 2020

Published online: 03 Sep 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article