A density-focused support vector data description method

Poovich Phaladiganon, Seoung Bum Kim, Victoria C P Chen

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

In novelty detection, support vector data description (SVDD) is a one-class classification technique that constructs a boundary to differentiate novel from normal patterns. However, boundaries constructed by SVDDdo not consider the density of the data. Data points located in low density regions are more likely to be novel patterns because they are remote from their neighbors. This study presents a density-focused SVDD (DFSVDD), for which its boundary considers both shape and the dense region of the data. Two distance measures, the kernel distance and the density distance, are combined to construct the DFSVDD boundary. The kernel distance can be obtained by solving a quadratic optimization, while support vectors are used to obtain the density distance. A simulation study was conducted to evaluate the performance of the proposed DFSVDD and was then compared with the traditional SVDD. The proposed method performed better than SVDD in terms of the area under the receiver operating characteristic curve.

Original languageEnglish
Pages (from-to)879-890
Number of pages12
JournalQuality and Reliability Engineering International
Volume30
Issue number6
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Data description

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Management Science and Operations Research

Cite this

A density-focused support vector data description method. / Phaladiganon, Poovich; Kim, Seoung Bum; Chen, Victoria C P.

In: Quality and Reliability Engineering International, Vol. 30, No. 6, 01.01.2014, p. 879-890.

Research output: Contribution to journalArticle

Phaladiganon, Poovich ; Kim, Seoung Bum ; Chen, Victoria C P. / A density-focused support vector data description method. In: Quality and Reliability Engineering International. 2014 ; Vol. 30, No. 6. pp. 879-890.
@article{76e0fb10989842fdab9573a91e840d97,
title = "A density-focused support vector data description method",
abstract = "In novelty detection, support vector data description (SVDD) is a one-class classification technique that constructs a boundary to differentiate novel from normal patterns. However, boundaries constructed by SVDDdo not consider the density of the data. Data points located in low density regions are more likely to be novel patterns because they are remote from their neighbors. This study presents a density-focused SVDD (DFSVDD), for which its boundary considers both shape and the dense region of the data. Two distance measures, the kernel distance and the density distance, are combined to construct the DFSVDD boundary. The kernel distance can be obtained by solving a quadratic optimization, while support vectors are used to obtain the density distance. A simulation study was conducted to evaluate the performance of the proposed DFSVDD and was then compared with the traditional SVDD. The proposed method performed better than SVDD in terms of the area under the receiver operating characteristic curve.",
keywords = "Data mining, Density, Novelty detection, One-class classification, Support vector data description",
author = "Poovich Phaladiganon and Kim, {Seoung Bum} and Chen, {Victoria C P}",
year = "2014",
month = "1",
day = "1",
doi = "10.1002/qre.1688",
language = "English",
volume = "30",
pages = "879--890",
journal = "Quality and Reliability Engineering International",
issn = "0748-8017",
publisher = "John Wiley and Sons Ltd",
number = "6",

}

TY - JOUR

T1 - A density-focused support vector data description method

AU - Phaladiganon, Poovich

AU - Kim, Seoung Bum

AU - Chen, Victoria C P

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In novelty detection, support vector data description (SVDD) is a one-class classification technique that constructs a boundary to differentiate novel from normal patterns. However, boundaries constructed by SVDDdo not consider the density of the data. Data points located in low density regions are more likely to be novel patterns because they are remote from their neighbors. This study presents a density-focused SVDD (DFSVDD), for which its boundary considers both shape and the dense region of the data. Two distance measures, the kernel distance and the density distance, are combined to construct the DFSVDD boundary. The kernel distance can be obtained by solving a quadratic optimization, while support vectors are used to obtain the density distance. A simulation study was conducted to evaluate the performance of the proposed DFSVDD and was then compared with the traditional SVDD. The proposed method performed better than SVDD in terms of the area under the receiver operating characteristic curve.

AB - In novelty detection, support vector data description (SVDD) is a one-class classification technique that constructs a boundary to differentiate novel from normal patterns. However, boundaries constructed by SVDDdo not consider the density of the data. Data points located in low density regions are more likely to be novel patterns because they are remote from their neighbors. This study presents a density-focused SVDD (DFSVDD), for which its boundary considers both shape and the dense region of the data. Two distance measures, the kernel distance and the density distance, are combined to construct the DFSVDD boundary. The kernel distance can be obtained by solving a quadratic optimization, while support vectors are used to obtain the density distance. A simulation study was conducted to evaluate the performance of the proposed DFSVDD and was then compared with the traditional SVDD. The proposed method performed better than SVDD in terms of the area under the receiver operating characteristic curve.

KW - Data mining

KW - Density

KW - Novelty detection

KW - One-class classification

KW - Support vector data description

UR - http://www.scopus.com/inward/record.url?scp=84908066205&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84908066205&partnerID=8YFLogxK

U2 - 10.1002/qre.1688

DO - 10.1002/qre.1688

M3 - Article

AN - SCOPUS:84908066205

VL - 30

SP - 879

EP - 890

JO - Quality and Reliability Engineering International

JF - Quality and Reliability Engineering International

SN - 0748-8017

IS - 6

ER -