Conference Proceedings

alpha-Information-theoretic Privacy Watch- dog and Optimal Privatization Scheme

Ni Ding, Mohammad Amin Zarrabian, Parastoo Sadeghi

2021 IEEE International Symposium on Information Theory (ISIT) | IEEE | Published : 2021

Abstract

This paper proposes an α -lift measure for data privacy and determines the optimal privatization scheme that minimizes the α -lift in the watchdog method. To release useful data X that is correlated with sensitive data S , the ratio of the posterior belief to the prior belief on sensitive data with respect to the useful data is called ‘lift’, which quantifies privacy risk. The α -lift denoted by ℓα(x) is proposed as the Lα -norm of the lift for a given realization x . This is a tunable measure: when α<∞ , each lift is weighted by its likelihood of appearing in the dataset (w.r.t. the marginal probability p(s) ); for α=∞, α -lift reduces to the existing maximum lift. To generate the sanitized..

View full abstract

University of Melbourne Researchers

Grants

Awarded by ARC Future Fellowship


Funding Acknowledgements

N. Ding is with the School of Computing and Information Systems, University of Melbourne, Melbourne. M. A. Zarrabian is with the ANU College of Engineering and Computer Science at the Australian National University. P. Sadeghi is with the School of Engineering and Information Technology, University of New South Wales, Canberra. The work of P. Sadeghi and M. A. Zarrabian is supported by the ARC Future Fellowship FT190100429 and partly by the Data61 CRP: IT-PPUB. Emails:ni.ding@unimelb.edu.au,u7139330@anu.edu.au,p.sadeghi@unsw.edu.au.