Privacy Preserving for High-Dimensional Data Using Overlapping Slicing

Paper Topic :

Data Mining

Author Name :

Zankhana Prajapati

Abstract :

Privacy preserving is one of the important method in data mining to hide the sensitive information. Anonymization techniques like generalization and bucketization are used for privacy preserving. In generalization loses some amount of information especially for high dimensional data and bucketization does not avoid membership disclosure. Slicing is a new approach to privacy preserving data publishing in which the data is partitioned horizontally and vertically. In Overlapping slicing there are three steps, attribute partitioning, Tuple partitioning and last column generalization. We use an efficient algorithm called chi_matrix for attribute correlation. For tuple partitioning we use more effective tuple grouping algorithm.

Download Article