Journal of Engineering and Applied Sciences
ISSN: Online 1818-7803ISSN: Print 1816-949x
Abstract
In recent years, the growing capacity of information storage devices has led to increased storing personal information about customers and individuals for various purposes. Data mining needs extensive amount of data to do analysis for finding out patterns and other information which could be helpful for business growth, tracking health data, improving services, etc. This information can be misused for many reasons like identity theft, fake credit/debit card transactions, etc. To avoid these situations, data mining techniques which secure privacy are proposed. Data perturbation, knowledge hiding, secure multiparty computation and privacy aware knowledge sharing are some of the techniques of privacy preserving data mining. A combination of these approaches is applied to get better privacy. In this study, we discuss in detail about geometric data perturbation technique and k-anonymization technique and prove that data mining results after perturbation and anonymization also are not changed much.
How to cite this article:
Naga Prasanthi Kundeti and M.V.P. Chandra Sekhara Rao. A Combined Approach for Privacy Preserving Classification Mining.
DOI: https://doi.org/10.36478/jeasci.2019.188.194
URL: https://www.makhillpublications.co/view-article/1816-949x/jeasci.2019.188.194