A new maximal-margin spherical-structured multi-class support vector machine

Pei Yi Hao, Jung-Hsien Chiang, Yen Hsiu Lin

Research output: Contribution to journalArticle

48 Citations (Scopus)

Abstract

Support vector machines (SVMs), initially proposed for two-class classification problems, have been very successful in pattern recognition problems. For multi-class classification problems, the standard hyperplane-based SVMs are made by constructing and combining several maximal-margin hyperplanes, and each class of data is confined into a certain area constructed by those hyperplanes. Instead of using hyperplanes, hyperspheres that tightly enclosed the data of each class can be used. Since the class-specific hyperspheres are constructed for each class separately, the spherical-structured SVMs can be used to deal with the multi-class classification problem easily. In addition, the center and radius of the class-specific hypersphere characterize the distribution of examples from that class, and may be useful for dealing with imbalance problems. In this paper, we incorporate the concept of maximal margin into the spherical-structured SVMs. Besides, the proposed approach has the advantage of using a new parameter on controlling the number of support vectors. Experimental results show that the proposed method performs well on both artificial and benchmark datasets.

Original languageEnglish
Pages (from-to)98-111
Number of pages14
JournalApplied Intelligence
Volume30
Issue number2
DOIs
Publication statusPublished - 2009 Apr 1

Fingerprint

Support vector machines
Pattern recognition

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Cite this

@article{8bb9aa5e165e4934b37cd267889d3ebb,
title = "A new maximal-margin spherical-structured multi-class support vector machine",
abstract = "Support vector machines (SVMs), initially proposed for two-class classification problems, have been very successful in pattern recognition problems. For multi-class classification problems, the standard hyperplane-based SVMs are made by constructing and combining several maximal-margin hyperplanes, and each class of data is confined into a certain area constructed by those hyperplanes. Instead of using hyperplanes, hyperspheres that tightly enclosed the data of each class can be used. Since the class-specific hyperspheres are constructed for each class separately, the spherical-structured SVMs can be used to deal with the multi-class classification problem easily. In addition, the center and radius of the class-specific hypersphere characterize the distribution of examples from that class, and may be useful for dealing with imbalance problems. In this paper, we incorporate the concept of maximal margin into the spherical-structured SVMs. Besides, the proposed approach has the advantage of using a new parameter on controlling the number of support vectors. Experimental results show that the proposed method performs well on both artificial and benchmark datasets.",
author = "Hao, {Pei Yi} and Jung-Hsien Chiang and Lin, {Yen Hsiu}",
year = "2009",
month = "4",
day = "1",
doi = "10.1007/s10489-007-0101-z",
language = "English",
volume = "30",
pages = "98--111",
journal = "Applied Intelligence",
issn = "0924-669X",
publisher = "Springer Netherlands",
number = "2",

}

A new maximal-margin spherical-structured multi-class support vector machine. / Hao, Pei Yi; Chiang, Jung-Hsien; Lin, Yen Hsiu.

In: Applied Intelligence, Vol. 30, No. 2, 01.04.2009, p. 98-111.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A new maximal-margin spherical-structured multi-class support vector machine

AU - Hao, Pei Yi

AU - Chiang, Jung-Hsien

AU - Lin, Yen Hsiu

PY - 2009/4/1

Y1 - 2009/4/1

N2 - Support vector machines (SVMs), initially proposed for two-class classification problems, have been very successful in pattern recognition problems. For multi-class classification problems, the standard hyperplane-based SVMs are made by constructing and combining several maximal-margin hyperplanes, and each class of data is confined into a certain area constructed by those hyperplanes. Instead of using hyperplanes, hyperspheres that tightly enclosed the data of each class can be used. Since the class-specific hyperspheres are constructed for each class separately, the spherical-structured SVMs can be used to deal with the multi-class classification problem easily. In addition, the center and radius of the class-specific hypersphere characterize the distribution of examples from that class, and may be useful for dealing with imbalance problems. In this paper, we incorporate the concept of maximal margin into the spherical-structured SVMs. Besides, the proposed approach has the advantage of using a new parameter on controlling the number of support vectors. Experimental results show that the proposed method performs well on both artificial and benchmark datasets.

AB - Support vector machines (SVMs), initially proposed for two-class classification problems, have been very successful in pattern recognition problems. For multi-class classification problems, the standard hyperplane-based SVMs are made by constructing and combining several maximal-margin hyperplanes, and each class of data is confined into a certain area constructed by those hyperplanes. Instead of using hyperplanes, hyperspheres that tightly enclosed the data of each class can be used. Since the class-specific hyperspheres are constructed for each class separately, the spherical-structured SVMs can be used to deal with the multi-class classification problem easily. In addition, the center and radius of the class-specific hypersphere characterize the distribution of examples from that class, and may be useful for dealing with imbalance problems. In this paper, we incorporate the concept of maximal margin into the spherical-structured SVMs. Besides, the proposed approach has the advantage of using a new parameter on controlling the number of support vectors. Experimental results show that the proposed method performs well on both artificial and benchmark datasets.

UR - http://www.scopus.com/inward/record.url?scp=62949089985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=62949089985&partnerID=8YFLogxK

U2 - 10.1007/s10489-007-0101-z

DO - 10.1007/s10489-007-0101-z

M3 - Article

VL - 30

SP - 98

EP - 111

JO - Applied Intelligence

JF - Applied Intelligence

SN - 0924-669X

IS - 2

ER -