Speaker
Dr
Makhamisa Senekane
(Department of Physics and Electronics, National University of Lesotho, Roma, Lesotho)
Description
The proliferation of data and data analysis has resulted in a need to pay more attention to the security of the the data being analyzed. Different options have been pursued to guarantee privacy of these statistical databases generated. However, most of such options cannot offer guarantee of security of the statistical databases. On the other hand, differential privacy is able to meet such a requirement of security definition. It guarantees privacy of an individual against an adversary with arbitrary auxiliary information. Differential privacy can be categorized as either &epsilon- or (&epsilon, &delta)-differential privacy. In this work, we report a mechanism for (&epsilon, &delta)-differential privacy using Student's t probability distribution. Our results guarantee the privacy and the utility of the mechanism, and this underlines the utility of this mechanism.
Supervisor details<br><b>If not a student, type N/A.</b><br>Student abstract submision<br>requires supervisor permission:<br>please give their name,<br> institution and email address.
N/A
| Consideration for<br>student awards<br><b>Choose one option<br>from those below.</b><br>N/A<br>Hons<br>MSc<br>PhD | N/A |
|---|---|
| Please confirm that you<br>have carefully read the<br>abstract submission instructions<br>under the menu item<br>"Call for Abstracts"<br><b>(Yes / No)</b> | Yes |
Primary authors
Prof.
Francesco Petruccione
(UKZN)
Dr
Makhamisa Senekane
(Department of Physics and Electronics, National University of Lesotho, Roma, Lesotho)
Dr
Molibeli Taele
(National University of Lesotho)
