Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data

Hwanjo Yu, Xiaoqian Jiang, Jaideep Vaidya

Research output: Chapter in Book/Report/Conference proceedingConference contribution

118 Citations (Scopus)

Abstract

Traditional Data Mining and Knowledge Discovery algorithms assume free access to data, either at a centralized location or in federated form. Increasingly, privacy and security concerns restrict this access, thus derailing data mining projects. What we need is distributed knowledge discovery that is sensitive to this problem. The key is to obtain valid results, while providing guarantees on the non-disclosure of data. Support vector machine classification is one of the most widely used classification methodologies in data mining and machine learning. It is based on solid theoretical foundations and has wide practical application. This paper proposes a privacy-preserving solution for support vector machine (SVM) classification, PP-SVM for short. Our solution constructs the global SVM classification model from the data distributed at multiple parties, without disclosing the data of each party to others. We assume that data is horizontally partitioned - each party collects the same features of information for different data objects. We quantify the security and efficiency of the proposed method, and highlight future challenges.

Original languageEnglish (US)
Title of host publicationApplied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing
Pages603-610
Number of pages8
StatePublished - Nov 21 2006
Event2006 ACM Symposium on Applied Computing - Dijon, France
Duration: Apr 23 2006Apr 27 2006

Publication series

NameProceedings of the ACM Symposium on Applied Computing
Volume1

Other

Other2006 ACM Symposium on Applied Computing
CountryFrance
CityDijon
Period4/23/064/27/06

Fingerprint

Support vector machines
Data mining
Learning systems

All Science Journal Classification (ASJC) codes

  • Software

Keywords

  • Privacy-preserving data mining
  • Support vector machine

Cite this

Yu, H., Jiang, X., & Vaidya, J. (2006). Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data. In Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing (pp. 603-610). (Proceedings of the ACM Symposium on Applied Computing; Vol. 1).
Yu, Hwanjo ; Jiang, Xiaoqian ; Vaidya, Jaideep. / Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data. Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing. 2006. pp. 603-610 (Proceedings of the ACM Symposium on Applied Computing).
@inproceedings{5a48c43129d340b49755cb4dc30cee22,
title = "Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data",
abstract = "Traditional Data Mining and Knowledge Discovery algorithms assume free access to data, either at a centralized location or in federated form. Increasingly, privacy and security concerns restrict this access, thus derailing data mining projects. What we need is distributed knowledge discovery that is sensitive to this problem. The key is to obtain valid results, while providing guarantees on the non-disclosure of data. Support vector machine classification is one of the most widely used classification methodologies in data mining and machine learning. It is based on solid theoretical foundations and has wide practical application. This paper proposes a privacy-preserving solution for support vector machine (SVM) classification, PP-SVM for short. Our solution constructs the global SVM classification model from the data distributed at multiple parties, without disclosing the data of each party to others. We assume that data is horizontally partitioned - each party collects the same features of information for different data objects. We quantify the security and efficiency of the proposed method, and highlight future challenges.",
keywords = "Privacy-preserving data mining, Support vector machine",
author = "Hwanjo Yu and Xiaoqian Jiang and Jaideep Vaidya",
year = "2006",
month = "11",
day = "21",
language = "English (US)",
isbn = "1595931082",
series = "Proceedings of the ACM Symposium on Applied Computing",
pages = "603--610",
booktitle = "Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing",

}

Yu, H, Jiang, X & Vaidya, J 2006, Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data. in Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing. Proceedings of the ACM Symposium on Applied Computing, vol. 1, pp. 603-610, 2006 ACM Symposium on Applied Computing, Dijon, France, 4/23/06.

Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data. / Yu, Hwanjo; Jiang, Xiaoqian; Vaidya, Jaideep.

Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing. 2006. p. 603-610 (Proceedings of the ACM Symposium on Applied Computing; Vol. 1).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data

AU - Yu, Hwanjo

AU - Jiang, Xiaoqian

AU - Vaidya, Jaideep

PY - 2006/11/21

Y1 - 2006/11/21

N2 - Traditional Data Mining and Knowledge Discovery algorithms assume free access to data, either at a centralized location or in federated form. Increasingly, privacy and security concerns restrict this access, thus derailing data mining projects. What we need is distributed knowledge discovery that is sensitive to this problem. The key is to obtain valid results, while providing guarantees on the non-disclosure of data. Support vector machine classification is one of the most widely used classification methodologies in data mining and machine learning. It is based on solid theoretical foundations and has wide practical application. This paper proposes a privacy-preserving solution for support vector machine (SVM) classification, PP-SVM for short. Our solution constructs the global SVM classification model from the data distributed at multiple parties, without disclosing the data of each party to others. We assume that data is horizontally partitioned - each party collects the same features of information for different data objects. We quantify the security and efficiency of the proposed method, and highlight future challenges.

AB - Traditional Data Mining and Knowledge Discovery algorithms assume free access to data, either at a centralized location or in federated form. Increasingly, privacy and security concerns restrict this access, thus derailing data mining projects. What we need is distributed knowledge discovery that is sensitive to this problem. The key is to obtain valid results, while providing guarantees on the non-disclosure of data. Support vector machine classification is one of the most widely used classification methodologies in data mining and machine learning. It is based on solid theoretical foundations and has wide practical application. This paper proposes a privacy-preserving solution for support vector machine (SVM) classification, PP-SVM for short. Our solution constructs the global SVM classification model from the data distributed at multiple parties, without disclosing the data of each party to others. We assume that data is horizontally partitioned - each party collects the same features of information for different data objects. We quantify the security and efficiency of the proposed method, and highlight future challenges.

KW - Privacy-preserving data mining

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=33751023839&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33751023839&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1595931082

SN - 9781595931085

T3 - Proceedings of the ACM Symposium on Applied Computing

SP - 603

EP - 610

BT - Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing

ER -

Yu H, Jiang X, Vaidya J. Privacy-preserving SVM using nonlinear kernels on horizontally partitioned data. In Applied Computing 2006 - The 21st Annual ACM Symposium on Applied Computing - Proceedings of the 2006 ACM Symposium on Applied Computing. 2006. p. 603-610. (Proceedings of the ACM Symposium on Applied Computing).