A New Method for Preserving Privacy in Data Publishing Against Attribute and Identity Disclosure Risk

Abstract

Sharing, transferring, mining and publishing data are fundamental operations in day to day life. Preserving the privacy of individuals is essential one. Sensitive personal information must be protected when data are published. There are two kinds of risks namely attribute disclosure and identity disclosure that affects privacy of individuals whose data are published. Early Researchers have contributed new methods namely k-anonymity, l-diversity, t-closeness to preserve privacy. K-anonymity method preserves privacy of individuals against identity disclosure attack alone. But Attribute disclosure attack makes compromise this method. Limitation of k-anonymity is fulfilled through l-diversity method. but it does not satisfies the privacy against identity disclosure attack and attribute disclosure attack in some scenarios. The efficiency of t-closeness method is better than k-anonymity and l-diversity. But the complexity of Computation is more than other proposed methods. In this paper, the authors contributes a new method for preserving the privacy of individuals’ sensitive information from attribute and identity disclosure attacks. In the proposed method, privacy preservation is full filled through generalization of quasi identifiers by setting range values.