Hybrid techniques for Privacy preserving in Data Mining
J.Paranthaman1, Dr. T Aruldoss Albert Victoire 2
1. Assistant Professor in Computer Science and Engineering, University College of Engineering,
Pattukottai Campus, Rajamadam- 614 701.
2. Associate Professor, Department of Electrical and Electronics Engineering, Regional Centre-Coimbatore,
Anna University, Coimbatore - 641 047.
Abstract: With the technology advancements, most of the corporations maintain their huge amount of electronic data in the databases and these databases are accessible using internet. These data are used by data miners to extract useful information. There is a threat to the privacy of the data while performing data mining tasks. Anonymization is one of the methods that transform actual data using generalization or suppression techniques, so that private information of individuals is masked. K-Anonymity transforms data into a set of equivalence classes and each class has a set of K- records indistinguishable from each other.In this proposed work, k-anonymity is used for privacy preserving while applying data mining algorithms. Hybrid technique simulated annealing with a genetic algorithm is used to optimize the feature selection. For evaluation, the mushroom data set anonymized to different levels for preserving privacy and hybrid technique for optimization is used. Experimental results demonstrate that the proposed method achieve satisfactory results.
[J.Paranthaman, Dr. T Aruldoss Albert Victoire. Hybrid techniques for Privacy preserving in Data Mining. Life Sci J 2013; 10(7s): 471-475] (ISSN:1097-8135). http://www.lifesciencesite.com. 73
Keywords: Privacy preserving data mining; k-anonymity; simulated annealing; Genetic algorithm