ATPM: Achieving t-Closeness using Particle Swarm Optimization and Movement of Record

Megha Bisht, Pranav Mangla, Sanya Saxena, Vartika Puri

Abstract: The amount of data available to us is increasing day by day. Data which is available online, especially the data from medical institutions, super markets and government has many uses like future research and innovation of new technologies, but this data may contain certain confidential information about a person like diagnosis details, purchase history etc. Privacy Preserving Data Publishing (PPDP) is a technique including privacy models like, k-anonymity, l-diversity and t-closeness that provide us with a scheme to publish data online without compromising the privacy of an individual. The purpose of this paper is to achieve t-closeness in such a way that the disclosure risk of patient?s data is minimized while maximizing the usefulness of the data for research purposes. This paper proposes an algorithm named as Achieving t-Closeness using Particle Swarm Optimization and Movement of record (ATPM) that achieves t-closeness in two phases. In the first phase, Particle Swarm Optimization is used to make equivalence classes. Second step involves movement of records from one equivalence class to other considering distribution of sensitive attributes in the equivalence classes to achieve t-closeness. The proposed work gives high privacy guarantee with low information loss.

Keywords: Privacy Preserving Data Publishing, T-closeness, Particle Swarm Optimization, Sensitive attribute