Optimal Univariate Microaggregation for Privacy Preservation in Data Mining

Abstract

In recent years, with the massive development in Internet, data collection and data warehousing technologies, privacy preservation has become one of the greater concerns in data mining. For this reason, several data mining algorithms integrating privacy preserving techniques have been developed in order to prevent the disclosure of sensitive information during the knowledge discovery. A number of effective methods for Privacy Preserving Data Mining (PPDM) have been proposed in the literature. In this paper, we present a brief introduction of different kinds of Microaggregation techniques with their merits and demerits and propose Optimal noise addition based Univariate Microaggregation for anonymizing the individual records. Through the experimental results, our proposed technique is validated to prevent the disclosure of sensitive data without degradation of data utilization. Our work highlights some discussions about future work and promising directions in the perspective of privacy preservation in data mining.