Isometric Relocation of Data by Sequencing of Sub-Clusters for Privacy Preservation in Data Mining

Privacy preservation in data mining is a pioneering research area as the security of increasing amount of data is under risks. Privacy preservation in Data Mining [PPDM] is a delicate task as there is a trade-off between data Anonymization and their utility. Existing PPDM techniques uses Anonymization using randomization, generalization or suppression which reduces the utility of data. They also do not work on the data mining parameters like correlation, centroids etc., This paper provides a solution to handle this trade-off in an efficient way using Isometric relocation. The work uses isometric relocation as it maintains the correlation and data mining results. The methodology is explained with the algorithm and its performance is compared using real-life datasets with existing techniques on various metrics after exhaustive experimentations.