A Robust Privacy Preservation by Combination of Additive and Multiplicative Data Perturbation for Privacy Preserving Data Mining

Abstract

of our daily activities are now routinely recorded and analysed by a variety of governmental and commercial organizations for the purpose of security and business related applications. From telephone calls to credit card purchases, from internet surfing to medical prescription refills, we generate data with almost every action we take. These data sets need to be analyzed for identifying patterns which can be used to predict future behaviour. However, data owners may not be willing to share the real values of their data due to privacy reason. Hence, some amount of privacy preservation needs to be done on data before it is released. Privacy preserving data mining (PPDM) tends to transform original data, so that sensitive data are preserved. In this paper we have proposed a new method CAMDP (Combination of Additive and Multiplicative Data Perturbation) for privacy preserving in data mining.