Multiclass Lagrangian support vector machine

Jae Pil Hwang; Baehoon Choi; In Wha Hong; Euntai Kim

doi:10.1007/s00521-011-0755-7

Multiclass Lagrangian support vector machine

Jae Pil Hwang, Baehoon Choi, In Wha Hong, Euntai Kim

Department of Electrical and Electronic Engineering

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

A support vector machine (SVM) has been developed for two-class problems, although its application to multiclass problems is not straightforward. This paper proposes a new Lagrangian SVM (LSVM) for application to multiclass problems. The multiclass Lagrangian SVM is formulated as a single optimization problem considering all the classes together, and a training method tailored to the multiclass problem is presented. A multiclass output representation matrix is defined to simplify the optimization formulation and associated training method. The proposed method is applied to some benchmark datasets in repository, and its effectiveness is demonstrated via simulation.

Original language	English
Pages (from-to)	703-710
Number of pages	8
Journal	Neural Computing and Applications
Volume	22
Issue number	3-4
DOIs	https://doi.org/10.1007/s00521-011-0755-7
Publication status	Published - 2013 Mar

Bibliographical note

Funding Information:
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science and Technology (2011-0005274).

All Science Journal Classification (ASJC) codes

Software
Artificial Intelligence

Access to Document

10.1007/s00521-011-0755-7

Cite this

@article{7c1282a30d224769bf8cae43de3b2d95,

title = "Multiclass Lagrangian support vector machine",

abstract = "A support vector machine (SVM) has been developed for two-class problems, although its application to multiclass problems is not straightforward. This paper proposes a new Lagrangian SVM (LSVM) for application to multiclass problems. The multiclass Lagrangian SVM is formulated as a single optimization problem considering all the classes together, and a training method tailored to the multiclass problem is presented. A multiclass output representation matrix is defined to simplify the optimization formulation and associated training method. The proposed method is applied to some benchmark datasets in repository, and its effectiveness is demonstrated via simulation.",

author = "Hwang, {Jae Pil} and Baehoon Choi and Hong, {In Wha} and Euntai Kim",

note = "Funding Information: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science and Technology (2011-0005274).",

year = "2013",

month = mar,

doi = "10.1007/s00521-011-0755-7",

language = "English",

volume = "22",

pages = "703--710",

journal = "Neural Computing and Applications",

issn = "0941-0643",

publisher = "Springer London",

number = "3-4",

}

TY - JOUR

T1 - Multiclass Lagrangian support vector machine

AU - Hwang, Jae Pil

AU - Choi, Baehoon

AU - Hong, In Wha

AU - Kim, Euntai

N1 - Funding Information: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science and Technology (2011-0005274).

PY - 2013/3

Y1 - 2013/3

N2 - A support vector machine (SVM) has been developed for two-class problems, although its application to multiclass problems is not straightforward. This paper proposes a new Lagrangian SVM (LSVM) for application to multiclass problems. The multiclass Lagrangian SVM is formulated as a single optimization problem considering all the classes together, and a training method tailored to the multiclass problem is presented. A multiclass output representation matrix is defined to simplify the optimization formulation and associated training method. The proposed method is applied to some benchmark datasets in repository, and its effectiveness is demonstrated via simulation.

AB - A support vector machine (SVM) has been developed for two-class problems, although its application to multiclass problems is not straightforward. This paper proposes a new Lagrangian SVM (LSVM) for application to multiclass problems. The multiclass Lagrangian SVM is formulated as a single optimization problem considering all the classes together, and a training method tailored to the multiclass problem is presented. A multiclass output representation matrix is defined to simplify the optimization formulation and associated training method. The proposed method is applied to some benchmark datasets in repository, and its effectiveness is demonstrated via simulation.

UR - http://www.scopus.com/inward/record.url?scp=84874022703&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874022703&partnerID=8YFLogxK

U2 - 10.1007/s00521-011-0755-7

DO - 10.1007/s00521-011-0755-7

M3 - Article

AN - SCOPUS:84874022703

VL - 22

SP - 703

EP - 710

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 0941-0643

IS - 3-4

ER -

Multiclass Lagrangian support vector machine

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this