File Download
 
Links for fulltext
(May Require Subscription)
 
Supplementary

Article: Discriminant analysis in pairwise kernel learning for SVM classification
  • Basic View
  • Metadata View
  • XML View
TitleDiscriminant analysis in pairwise kernel learning for SVM classification
 
AuthorsJiang, H1
Ching, WK1
Chu, D2
 
KeywordsClassification
Discriminant analysis
Kernel learning
Support vector machine
SVM
 
Issue Date2012
 
PublisherInderscience Publishers. The Journal's web site is located at http://www.inderscience.com/ijbra
 
CitationInternational Journal of Bioinformatics Research and Applications, 2012, v. 8 n. 3-4, p. 305-321 [How to Cite?]
DOI: http://dx.doi.org/10.1504/IJBRA.2012.048963
 
AbstractMultiple kernel learning arises when different types of kernels are employed simultaneously. In particular, in the situation that the data are from heterogeneous sources. In this study, we developed a new framework for determining the coefficients in learning pairwise kernels for classification in Support Vector Machines (SVM). The effectiveness of the proposed method was then demonstrated through the prediction of self-renewal and pluripotency mESCs stemness membership genes. It was also tested on the power of discrimination in DNA repair gene data. The promising formulation in learning coefficients for pairwise kernel learning was shown via experimental evaluation. This may provide a novel perspective for kernel learning in future applications.
 
ISSN1744-5485
2012 SCImago Journal Rankings: 0.189
 
DOIhttp://dx.doi.org/10.1504/IJBRA.2012.048963
 
DC FieldValue
dc.contributor.authorJiang, H
 
dc.contributor.authorChing, WK
 
dc.contributor.authorChu, D
 
dc.date.accessioned2012-09-20T07:56:20Z
 
dc.date.available2012-09-20T07:56:20Z
 
dc.date.issued2012
 
dc.description.abstractMultiple kernel learning arises when different types of kernels are employed simultaneously. In particular, in the situation that the data are from heterogeneous sources. In this study, we developed a new framework for determining the coefficients in learning pairwise kernels for classification in Support Vector Machines (SVM). The effectiveness of the proposed method was then demonstrated through the prediction of self-renewal and pluripotency mESCs stemness membership genes. It was also tested on the power of discrimination in DNA repair gene data. The promising formulation in learning coefficients for pairwise kernel learning was shown via experimental evaluation. This may provide a novel perspective for kernel learning in future applications.
 
dc.description.natureLink_to_subscribed_fulltext
 
dc.identifier.citationInternational Journal of Bioinformatics Research and Applications, 2012, v. 8 n. 3-4, p. 305-321 [How to Cite?]
DOI: http://dx.doi.org/10.1504/IJBRA.2012.048963
 
dc.identifier.doihttp://dx.doi.org/10.1504/IJBRA.2012.048963
 
dc.identifier.epage321
 
dc.identifier.hkuros208785
 
dc.identifier.issn1744-5485
2012 SCImago Journal Rankings: 0.189
 
dc.identifier.issue3-4
 
dc.identifier.pmid22961457
 
dc.identifier.scopuseid_2-s2.0-84866246773
 
dc.identifier.spage305
 
dc.identifier.urihttp://hdl.handle.net/10722/164183
 
dc.identifier.volume8
 
dc.languageeng
 
dc.publisherInderscience Publishers. The Journal's web site is located at http://www.inderscience.com/ijbra
 
dc.publisher.placeUnited Kingdom
 
dc.relation.ispartofInternational Journal of Bioinformatics Research and Applications
 
dc.rightsInternational Journal of Bioinformatics Research and Applications. Copyright © Inderscience Publishers.
 
dc.subjectClassification
 
dc.subjectDiscriminant analysis
 
dc.subjectKernel learning
 
dc.subjectSupport vector machine
 
dc.subjectSVM
 
dc.titleDiscriminant analysis in pairwise kernel learning for SVM classification
 
dc.typeArticle
 
<?xml encoding="utf-8" version="1.0"?>
<item><contributor.author>Jiang, H</contributor.author>
<contributor.author>Ching, WK</contributor.author>
<contributor.author>Chu, D</contributor.author>
<date.accessioned>2012-09-20T07:56:20Z</date.accessioned>
<date.available>2012-09-20T07:56:20Z</date.available>
<date.issued>2012</date.issued>
<identifier.citation>International Journal of Bioinformatics Research and Applications, 2012, v. 8 n. 3-4, p. 305-321</identifier.citation>
<identifier.issn>1744-5485</identifier.issn>
<identifier.uri>http://hdl.handle.net/10722/164183</identifier.uri>
<description.abstract>Multiple kernel learning arises when different types of kernels are employed simultaneously. In particular, in the situation that the data are from heterogeneous sources. In this study, we developed a new framework for determining the coefficients in learning pairwise kernels for classification in Support Vector Machines (SVM). The effectiveness of the proposed method was then demonstrated through the prediction of self-renewal and pluripotency mESCs stemness membership genes. It was also tested on the power of discrimination in DNA repair gene data. The promising formulation in learning coefficients for pairwise kernel learning was shown via experimental evaluation. This may provide a novel perspective for kernel learning in future applications.</description.abstract>
<language>eng</language>
<publisher>Inderscience Publishers. The Journal&apos;s web site is located at http://www.inderscience.com/ijbra</publisher>
<relation.ispartof>International Journal of Bioinformatics Research and Applications</relation.ispartof>
<rights>International Journal of Bioinformatics Research and Applications. Copyright &#169; Inderscience Publishers.</rights>
<subject>Classification</subject>
<subject>Discriminant analysis</subject>
<subject>Kernel learning</subject>
<subject>Support vector machine</subject>
<subject>SVM</subject>
<title>Discriminant analysis in pairwise kernel learning for SVM classification</title>
<type>Article</type>
<description.nature>Link_to_subscribed_fulltext</description.nature>
<identifier.doi>10.1504/IJBRA.2012.048963</identifier.doi>
<identifier.pmid>22961457</identifier.pmid>
<identifier.scopus>eid_2-s2.0-84866246773</identifier.scopus>
<identifier.hkuros>208785</identifier.hkuros>
<identifier.volume>8</identifier.volume>
<identifier.issue>3-4</identifier.issue>
<identifier.spage>305</identifier.spage>
<identifier.epage>321</identifier.epage>
<publisher.place>United Kingdom</publisher.place>
</item>
Author Affiliations
  1. The University of Hong Kong
  2. National University of Singapore