A hybrid data transformation approach for privacy preserving clustering of categorical data

Abstract

In today's information age there is a large availability of repositories storing various types of information about individuals. Data mining technology has emerged as a means of identifying patterns and trends from large quantities of data. The application of data mining technology to identify interesting patterns from these repositories leads to serious privacy concerns. Despite its benefit in a wide range of applications, data mining techniques also have raised a number of ethical issues. Some such issues are privacy, data security, intellectual property rights and many others. In this paper, we address the privacy problem against unauthorized secondary use of information. We focus primarily on privacy preserving data clustering on categorical data. In the proposed method, the categorical data is converted into binary data and it is transformed using geometric data transformation method. Then, clustering using conventional clustering algorithm is done on the transformed data to ensure privacy. © 2007 Springer.