Preventing interval-based inference by random data perturbation

Authors:
Yingjiu Li;Lingyu Wang;Sushil Jajodia
Affiliations:
Center for Secure Information Systems, George Mason University, Fairfax, VA;Center for Secure Information Systems, George Mason University, Fairfax, VA;Center for Secure Information Systems, George Mason University, Fairfax, VA
Venue:
PET'02 Proceedings of the 2nd international conference on Privacy enhancing technologies
Year:
2002

Citing 7
Cited 6

Security-control methods for statistical databases: a comparative study

ACM Computing Surveys (CSUR)
A modified random perturbation method for database security

ACM Transactions on Database Systems (TODS)
Security of random data perturbation methods

ACM Transactions on Database Systems (TODS)
Data Security

ACM Computing Surveys (CSUR)
The statistical security of a statistical database

ACM Transactions on Database Systems (TODS)
Security of Statistical Databases - Compromise through Attribute Correlational Modeling

Proceedings of the Second International Conference on Data Engineering
Inference Controls for Statistical Databases

Computer

Auditing Interval-Based Inference

CAiSE '02 Proceedings of the 14th International Conference on Advanced Information Systems Engineering
A Data Disclosure Policy for Count Data Based on the COM-Poisson Distribution

Management Science
Parity-based inference control for multi-dimensional range sum queries

Journal of Computer Security
An efficient online auditing approach to limit private data disclosure

Proceedings of the 12th International Conference on Extending Database Technology: Advances in Database Technology
Preventing range disclosure in k-anonymised data

Expert Systems with Applications: An International Journal
Disclosure analysis for two-way contingency tables

PSD'06 Proceedings of the 2006 CENEX-SDC project international conference on Privacy in Statistical Databases

Quantified Score

Hi-index	0.00

Visualization

Abstract

Random data perturbation (RDP) method is often used in statistical databases to prevent inference of sensitive information about individuals from legitimate sum queries. In this paper, we study the RDP method for preventing an important type of inference: interval-based inference. In terms of interval-based inference, the sensitive information about individuals is said to be compromised if an accurate enough interval, called inference interval, is obtained into which the value of the sensitive information must fall. We show that the RDP methods proposed in the literature are not effective for preventing such interval-based inference. Based on a new type of random distribution, called Ɛ-Gaussian distribution, we propose a new RDP method to guarantee no interval-based inference.