A Privacy-Preserving Data Mining Method Based on Singular Value Decomposition and Independent Component Analysis
By: Guang Li and Yadong Wang
Abstract
Privacy protection is indispensable in data mining, and many privacy-preserving data mining (PPDM) methods have been proposed. One such method is based on singular value decomposition (SVD), which uses SVD to find unimportant information for data mining and removes it to protect privacy. Independent component analysis (ICA) is another data analysis method. If both SVD and ICA are used, unimportant information can be extracted more comprehensively. Accordingly, this paper proposes a new PPDM method using both SVD and ICA. Experiments show that our method performs better in preserving privacy than the SVD-based methods while also maintaining data utility.
DOI: https://doi.org/10.2481/dsj.009-025 | Journal eISSN: 1683-1470
Language: English
Page range: 124 - 132
Published on: Feb 8, 2011
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year
Keywords:
© 2011 Guang Li, Yadong Wang, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.