Data Privacy Using Association Rule Mining and Gaussian Position

Project Discription

Abstract

As the main aim of the privacy preserving mining is to maintain the valuable information at the same time sensitive information remain unaffected. In this paper sensitive information of the dataset is perturbed by finding the patterns which are more crossing the minimum support value. Here Aprior algorithm is use for generating patterns. By using Gaussian function random session in the dataset are perturbed. Perturbation in this work was so done that size of the dataset remains same. Here proposed work has developed a de-perturbation algorithm for data regeneration as well. As proposed work takes care an size of dataset remain unaffected after applying perturbation algorithm. One more advantage changes were done randomly by using Gaussian increase the confusion of the intruder. Data owner can de-perturb the dataset in original form without any loss is also done in this work. ‘Results shows that proposed work has improve perturbation percentage, while execution time and data loss got decrease.

Gaussian Function

Let G₁ through G_L be L Gaussian random variables. They are said to be Gaussian if and only if each of them is a linear combination of multiple independent Gaussian random variables [10]. Equivalently, G₁ through G_L are jointly Gaussian if and only if any linear combination of them is also a Gaussian random variable. A vector formed by jointly Gaussian random variables is called a jointly Gaussian vector. 

Project Sample Image