The innovative secrecy measure for data broadcasting
Geeta Rani, R. Murugeswari, M Sakthimohan · 2017
Data mining is the development of analyzing files since different perspectives and shorting it into useful information. It is also called data or knowledge detection data. The k-anonymity which does not defend against from attributes disclosure. The notations of l-diversity which has amount of restrictions and also not guard against from attribute disclosure. Several scholars used a confidentiality notation called “closeness” but its base ideal `t' closeness needs thoughtful element must nearby to scattering attribute in total table so it is a main problematic. By using the EMD distance measure, the derivation is too hard for every measurement one has to choose ground distance between couples of sensitive attribute value so the suppression of full record is unnecessary during anonymization. In k-anonymity and in l-diversity sensitive attribute is detached so level of privacy is decreased. A new technique is established by combining generalization and suppression by using in all these three techniques for achieving a well data quality. In this paper, our desiderata are using the UC Irvine device knowledge repository a data from adult database which is composed from US census.