Privacy-Preserving Data Publishing Based on De-clustering

In recent years, privacy preservation has become a serious concern in publication of personal data because of the wide availability of personal data. In the literature, we know that the degree of privacy protection is really determined by the number of distinct sensitive values in each group which is classified according to quasi-identifiers. In this paper, we present a novel method to protect data privacy by partitioning the microdata into some groups based on de-clustering. In this method, we make the records contained in each group possess distinct sensitive values and ensure that the size of the minimal groups not to be less than a threshold zeta. According to a novel privacy measure proposed in this paper, our method can provide strong privacy protection. Extensive experiments confirm that our method can provide stronger privacy protection than the methods based on l-diversity.