Abstract
In this thesis, execute a machine learning approach for elegant edges exploitation differential privacy. Privacy preservation is important for machine learning and massive data processing, however, measures designed to guard personal data usually end in a substitution reduced utility of the coaching samples. These papers introduce a privacy conserving approach that will be applied to call tree learning, without a connected loss of accuracy. It describes an approach to the preservation of the privacy of collecting information samples in cases wherever data from the sample information has been lost partly. This approach converts the initial sample knowledge set into a group of Non-Sensitive information sets, from that the initial samples may not be rebuild without the complete group of unreal information sets. Meanwhile, a correct analysis is often designed directly from those unreal information sets. This novel approach may be applied directly to the information storage as shortly as the initial sample is collected.
Keywords
Data mining
Privacy Preserving
OTP
OBP
GP
Sources Anonymity
Cryptography Model.
Authors
How to Cite this Article
S. Dhivya, K.M. Padma priya (2019).
"EFFECTIVE SECURE MULTIVARIANT DATA PUBLICATION MODEL FOR CLOUD USING HIGH DIMENSIONAL CLASSIFICATION TREE ALGORITHM".
International Journal of Contemporary Research in Computer Science and Technology,
5(3), pp. 25-30.
