Differential Privacy Preservation Mechanism Using Bernstein Polynomial Function For Heart Disease Dataset

Document Type : Primary Research paper

Authors

1 Sri Krishna College of Engineering and Technology, Coimbatore, Tamil Nadu

2 Bannari Amman Institute of Technology, Sathyamangalam, Tamil Nadu

Abstract

Information is put away in various frameworks as emerging technologies allows
influential data gathering and processing. Protection and security have turn into a
longstanding challenging issue with advances in data and communication innovation.
Privacy preserving data mining makes the customer information more secure by means of
data perturbation and also it makes it harder to identify a person in an occurrence of data
is spilled. Machine learning has got attention recently due to an energetic advancement of
differential privacy (DP). DP is golden response scheme to address the privacy protection
in analysis of data but it is quite hard to implement on real world data. The proposed
system uses Bernstein polynomial function under differential privacy for perturbation.
Heart disease dataset is used in this work to analyze the performance between the original
and the modified dataset using DP with the classifier models decision tree, linear model,
random forest, SVM, linear model and neural network. The experiment results show the
minor variations in the accuracy, sensitivity and specificity measures.

Keywords

Int. J. of Aquatic Science
Volume 12, Issue 3 - Serial Number 3
ICMMNT-2021 International Virtual Conference on Materials, Manufacturing and Nanotechnology, 30th June, 2021.
June 2021
Pages 1662-1671

Files

Share

How to cite

Statistics