Knowledge and Information Systems

, Volume 28, Issue 2, pp 251–282

COAT: COnstraint-based anonymization of transactions

  • Grigorios Loukides
  • Aris Gkoulalas-Divanis
  • Bradley Malin
Regular Paper

DOI: 10.1007/s10115-010-0354-4

Cite this article as:
Loukides, G., Gkoulalas-Divanis, A. & Malin, B. Knowl Inf Syst (2011) 28: 251. doi:10.1007/s10115-010-0354-4

Abstract

Publishing transactional data about individuals in an anonymous form is increasingly required by organizations. Recent approaches ensure that potentially identifying information cannot be used to link published transactions to individuals’ identities. However, these approaches are inadequate to anonymize data that is both protected and practically useful in applications because they incorporate coarse privacy requirements, do not integrate utility requirements, and tend to explore a small portion of the solution space. In this paper, we propose the first approach for anonymizing transactional data under application-specific privacy and utility requirements. We model such requirements as constraints, investigate how these constraints can be specified, and propose COnstraint-based Anonymization of Transactions, an algorithm that anonymizes transactions using a flexible anonymization scheme to meet the specified constraints. Experiments with benchmark datasets verify that COAT significantly outperforms the current state-of-the-art algorithm in terms of data utility, while being comparable in terms of efficiency. Our approach is also shown to be effective in preserving both privacy and utility in a real-world scenario that requires disseminating patients’ information.

Keywords

Privacy-preserving data publication Anonymity Transactional data Context-aware anonymization Privacy constraints Utility constraints 

Copyright information

© Springer-Verlag London Limited 2010

Authors and Affiliations

  • Grigorios Loukides
    • 1
  • Aris Gkoulalas-Divanis
    • 2
  • Bradley Malin
    • 1
  1. 1.Department of Biomedical InformaticsVanderbilt UniversityNashvilleUSA
  2. 2.IBM Research-ZurichRüschlikonSwitzerland

Personalised recommendations