A General Framework for Privacy Preserving Sequential Data Publishing

In this paper, we study the problem of privacy preserving sequential data publishing for microdata table. For example, a hospital might release patient's records in every three months. An adversary may disclose the confidential information of an individual across different publications of data sets by linking quasi-identifier attributes associated with the sensitive values. Most of the published work reduces the data utility to prevent the linking attack on published data set. In this paper, we propose a general framework and algorithm that can handle sequential data publishing issues and can protect the published data set from the linking attack. The proposed sequential algorithm satisfies l-diversity and increases the data utility during data publication. Experimental results show that the proposed framework counter the published dataset from linking attack and keep more data utility than the existing methods.