Privacy Preservation for Continuous Decremental Data Publishing

Abstract

The privacy violation is a serious issue that must be considered when datasets are published to the outside scope of the data-collecting organization for gaining more data utility. To address this issue, a well-known privacy preservation model, (α, k)-Anonymity, is proposed. Unfortunately, this privacy preservation model is generally proposed to address privacy violation issues in datasets that are focused on performing one-time data publishing. Therefore, if published datasets are dynamic, i.e., the data of them can be always changed and performing multiple time data publishing such as continuous decremental datasets, (α, k)-Anonymity could be insufficient. For this reason, the vulnerabilities of (α, k)-Anonymity in continuous decremental datasets are identified in this work. Moreover, an exhaustive decremental privacy preservation algorithm for addressing privacy violation issues in continuous decremental datasets is also proposed. The aim of the proposed algorithm is that aside from privacy preservation constraints, the data utility of datasets is also maintained as much as possible. Furthermore, the experimental results are shown which indicate that the proposed algorithm is highly effective.