Assessing Privacy Using the Area Under the Receiver-Operator Characteristic Curve

• Activity Number: 402
• Type: Topic Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Gregory J. Matthews*+ and Ofer Harel and Robert H. Aseltine
• Companies: University of Connecticut and University of Connecticut and University of Connecticut Health Center
• Keywords: Privacy ; Confidentiality

Abstract

One of the most pressing issues in the confidentiality literature is the quantification of privacy. One proposal, $\epsilon$-differential privacy, has been set forth; however, the selection of an appropriate $\epsilon$ is difficult because its interpretation is unclear. The aim of this work is to provide a measure that is more easily interpreted and to allow for policy debate pertaining to how much privacy is acceptable. Our proposal sets the problem in a hypothesis testing framework and uses the area under the receiver-operator characteristic (ROC) curve as a measure of privacy. This method allows privacy to be compared using one scale, regardless of the privacy preserving mechanism. It also provides an easily interpretable privacy metric.