An experimental study of matrix-based data distortion methods

Jie Wang; Hualing Liu; Guangwei Hu; Jun Zhang; James M. Grogan

doi:10.1109/ICCIS.2010.328

An experimental study of matrix-based data distortion methods

Jie Wang, Hualing Liu, Guangwei Hu, Jun Zhang, James M. Grogan

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

A number of matrix-based data distortion methods are presented and experimentally studied in this paper. The performances of seven methods are compared in terms of utility, privacy and computational cost. We find that left multiplication based random projection methods are useless in data privacy protection. Even though there is no application-free solution in data privacy protection, the nonnegative matrix factorization (NMF) based method has an appealing privacy performance under the promise of a reasonable utility and computational cost. While the random projection method with a right multiplication of an orthogonal random matrix does well in support vector machine classification, its computational disadvantages may make it less attractive for an online analysis and processing application.

Original language	English
Title of host publication	Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010
Pages	952-955
Number of pages	4
DOIs	https://doi.org/10.1109/ICCIS.2010.328
State	Published - 2010
Event	2010 International Conference on Computational and Information Sciences, ICCIS2010 - Chengdu, Sichuan, China Duration: Dec 17 2010 → Dec 19 2010

Publication series

Name	Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010

Conference

Conference	2010 International Conference on Computational and Information Sciences, ICCIS2010
Country/Territory	China
City	Chengdu, Sichuan
Period	12/17/10 → 12/19/10

Keywords

Data distortion
Matrix decomposition
Privacy
Random projection

ASJC Scopus subject areas

Computational Theory and Mathematics
Information Systems

Access to Document

10.1109/ICCIS.2010.328

Cite this

Wang, J., Liu, H., Hu, G., Zhang, J., & Grogan, J. M. (2010). An experimental study of matrix-based data distortion methods. In Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010 (pp. 952-955). Article 5709415 (Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010). https://doi.org/10.1109/ICCIS.2010.328

@inproceedings{2eafca0166be4d34817db8b6ed7790c9,

title = "An experimental study of matrix-based data distortion methods",

abstract = "A number of matrix-based data distortion methods are presented and experimentally studied in this paper. The performances of seven methods are compared in terms of utility, privacy and computational cost. We find that left multiplication based random projection methods are useless in data privacy protection. Even though there is no application-free solution in data privacy protection, the nonnegative matrix factorization (NMF) based method has an appealing privacy performance under the promise of a reasonable utility and computational cost. While the random projection method with a right multiplication of an orthogonal random matrix does well in support vector machine classification, its computational disadvantages may make it less attractive for an online analysis and processing application.",

keywords = "Data distortion, Matrix decomposition, Privacy, Random projection",

author = "Jie Wang and Hualing Liu and Guangwei Hu and Jun Zhang and Grogan, {James M.}",

year = "2010",

doi = "10.1109/ICCIS.2010.328",

language = "English",

isbn = "9780769542706",

series = "Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010",

pages = "952--955",

booktitle = "Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010",

note = "2010 International Conference on Computational and Information Sciences, ICCIS2010 ; Conference date: 17-12-2010 Through 19-12-2010",

}

Wang, J, Liu, H, Hu, G, Zhang, J & Grogan, JM 2010, An experimental study of matrix-based data distortion methods. in Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010., 5709415, Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010, pp. 952-955, 2010 International Conference on Computational and Information Sciences, ICCIS2010, Chengdu, Sichuan, China, 12/17/10. https://doi.org/10.1109/ICCIS.2010.328

An experimental study of matrix-based data distortion methods. / Wang, Jie; Liu, Hualing; Hu, Guangwei et al.
Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010. 2010. p. 952-955 5709415 (Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An experimental study of matrix-based data distortion methods

AU - Wang, Jie

AU - Liu, Hualing

AU - Hu, Guangwei

AU - Zhang, Jun

AU - Grogan, James M.

PY - 2010

Y1 - 2010

N2 - A number of matrix-based data distortion methods are presented and experimentally studied in this paper. The performances of seven methods are compared in terms of utility, privacy and computational cost. We find that left multiplication based random projection methods are useless in data privacy protection. Even though there is no application-free solution in data privacy protection, the nonnegative matrix factorization (NMF) based method has an appealing privacy performance under the promise of a reasonable utility and computational cost. While the random projection method with a right multiplication of an orthogonal random matrix does well in support vector machine classification, its computational disadvantages may make it less attractive for an online analysis and processing application.

AB - A number of matrix-based data distortion methods are presented and experimentally studied in this paper. The performances of seven methods are compared in terms of utility, privacy and computational cost. We find that left multiplication based random projection methods are useless in data privacy protection. Even though there is no application-free solution in data privacy protection, the nonnegative matrix factorization (NMF) based method has an appealing privacy performance under the promise of a reasonable utility and computational cost. While the random projection method with a right multiplication of an orthogonal random matrix does well in support vector machine classification, its computational disadvantages may make it less attractive for an online analysis and processing application.

KW - Data distortion

KW - Matrix decomposition

KW - Privacy

KW - Random projection

UR - http://www.scopus.com/inward/record.url?scp=79952435961&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952435961&partnerID=8YFLogxK

U2 - 10.1109/ICCIS.2010.328

DO - 10.1109/ICCIS.2010.328

M3 - Conference contribution

AN - SCOPUS:79952435961

SN - 9780769542706

T3 - Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010

SP - 952

EP - 955

BT - Proceedings - 2010 International Conference on Computational and Information Sciences, ICCIS 2010

T2 - 2010 International Conference on Computational and Information Sciences, ICCIS2010

Y2 - 17 December 2010 through 19 December 2010

ER -

An experimental study of matrix-based data distortion methods

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this