Privacy Preserving Data Mining Based on Geometrical Data Transformation Method (GDTM) and K-Means Clustering Algorithm

Abstract

In current era of sharing unlimited digital information via the network, protecting the privacy of information is crucial even during the data mining process due to a high possibility of the information security risks such as being abused or leakage. Such problems motivate the research in Privacy Preserving Data Mining (PPDM) and it became one of the newest trends. Therefore, this papers reviews the related works in terms of issues, approaches, techniques, performance quantification as well as thorough discussions on pros and cons of previous researches. We also propose an improved PPDM that applying Geometrical Data Transformation Method (GDTM) and K-Means Clustering Algorithm for optimum accuracy of mining and zero data loss while preserving the privacy of information.