Privacy preservation for re-publication data by using probabilistic graph

Abstract

With the dynamism of data intensive applications, data can be changed by the insert, update, and delete operations, at all times. Thus, the privacy models are designed to protect the static dataset might not be able to cope with the case of the dynamic dataset effectively. m-invariance and m-distinct models are the well-known anonymization model which are proposed to protect the privacy data in the dynamic dataset. However, in their counting-based model, the privacy data of the target user could still be revealed on internally or fully updated datasets when they are analyzed using updated probability graph. In this paper, we propose a new privacy model for dynamic data publishing based on probability graph. Subsequently, in order to study the characteristics of the problem, we propose a brute-force algorithm to preserve the privacy and maintain the data quality. From the experiment results, our proposed model can guarantee the minimum probability of inferencing sensitive value.