A Review of Various Multiple Numerical and Categorical Sensitive Attribute for Preserving Privacy

Abstract

Nowadays Privacy Preserving in Data Publishing (PPDP) has rapidly growth into the research contents in data protection field and also it has rapidly grown in publication of personal data. Recent years in data publishing is exceptionally analytical, because, how to economically preserve numerical and categorical sensitive attribute. However, different tender have been designed for privacy preserving to protect numerical sensitive attribute in data publishing, Like k-anonymity, l-diversity, t-closeness, (epsilon, m)-anonymity another methods are implemented for protecting the privacy of data provider. In this paper, we propose a review of various multiple numerical and categorical sensitive attribute for preserving privacy and survey current existing techniques. Yet, we discuss the future instructions of privacy preserving in data publishing, that is we suggest modern technique for “A new enhanced slicing method for both numerical and categorical sensitive attribute via advanced clustering algorithm†that improving the privacy with less information loss, membership, attribute , identity disclosure and error ratio and also calculate distance between two attributes for categorical attribute. This technique suitable for both categorical and numerical sensitive attributes.