A new method combining LDA and PLS for dimension reduction.

Linear discriminant analysis (LDA) is a classical statistical approach for dimensionality reduction and classification. In many cases, the projection direction of the classical and extended LDA methods is not considered optimal for special applications. Herein we combine the Partial Least Squares (P...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Liang Tang, Silong Peng, Yiming Bi, Peng Shan, Xiyuan Hu
Formato: article
Lenguaje:EN
Publicado: Public Library of Science (PLoS) 2014
Materias:
R
Q
Acceso en línea:https://doaj.org/article/da1f0ba2699846309ca5312911630bc9
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:da1f0ba2699846309ca5312911630bc9
record_format dspace
spelling oai:doaj.org-article:da1f0ba2699846309ca5312911630bc92021-11-18T08:19:42ZA new method combining LDA and PLS for dimension reduction.1932-620310.1371/journal.pone.0096944https://doaj.org/article/da1f0ba2699846309ca5312911630bc92014-01-01T00:00:00Zhttps://www.ncbi.nlm.nih.gov/pmc/articles/pmid/24820185/pdf/?tool=EBIhttps://doaj.org/toc/1932-6203Linear discriminant analysis (LDA) is a classical statistical approach for dimensionality reduction and classification. In many cases, the projection direction of the classical and extended LDA methods is not considered optimal for special applications. Herein we combine the Partial Least Squares (PLS) method with LDA algorithm, and then propose two improved methods, named LDA-PLS and ex-LDA-PLS, respectively. The LDA-PLS amends the projection direction of LDA by using the information of PLS, while ex-LDA-PLS is an extension of LDA-PLS by combining the result of LDA-PLS and LDA, making the result closer to the optimal direction by an adjusting parameter. Comparative studies are provided between the proposed methods and other traditional dimension reduction methods such as Principal component analysis (PCA), LDA and PLS-LDA on two data sets. Experimental results show that the proposed method can achieve better classification performance.Liang TangSilong PengYiming BiPeng ShanXiyuan HuPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 9, Iss 5, p e96944 (2014)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Liang Tang
Silong Peng
Yiming Bi
Peng Shan
Xiyuan Hu
A new method combining LDA and PLS for dimension reduction.
description Linear discriminant analysis (LDA) is a classical statistical approach for dimensionality reduction and classification. In many cases, the projection direction of the classical and extended LDA methods is not considered optimal for special applications. Herein we combine the Partial Least Squares (PLS) method with LDA algorithm, and then propose two improved methods, named LDA-PLS and ex-LDA-PLS, respectively. The LDA-PLS amends the projection direction of LDA by using the information of PLS, while ex-LDA-PLS is an extension of LDA-PLS by combining the result of LDA-PLS and LDA, making the result closer to the optimal direction by an adjusting parameter. Comparative studies are provided between the proposed methods and other traditional dimension reduction methods such as Principal component analysis (PCA), LDA and PLS-LDA on two data sets. Experimental results show that the proposed method can achieve better classification performance.
format article
author Liang Tang
Silong Peng
Yiming Bi
Peng Shan
Xiyuan Hu
author_facet Liang Tang
Silong Peng
Yiming Bi
Peng Shan
Xiyuan Hu
author_sort Liang Tang
title A new method combining LDA and PLS for dimension reduction.
title_short A new method combining LDA and PLS for dimension reduction.
title_full A new method combining LDA and PLS for dimension reduction.
title_fullStr A new method combining LDA and PLS for dimension reduction.
title_full_unstemmed A new method combining LDA and PLS for dimension reduction.
title_sort new method combining lda and pls for dimension reduction.
publisher Public Library of Science (PLoS)
publishDate 2014
url https://doaj.org/article/da1f0ba2699846309ca5312911630bc9
work_keys_str_mv AT liangtang anewmethodcombiningldaandplsfordimensionreduction
AT silongpeng anewmethodcombiningldaandplsfordimensionreduction
AT yimingbi anewmethodcombiningldaandplsfordimensionreduction
AT pengshan anewmethodcombiningldaandplsfordimensionreduction
AT xiyuanhu anewmethodcombiningldaandplsfordimensionreduction
AT liangtang newmethodcombiningldaandplsfordimensionreduction
AT silongpeng newmethodcombiningldaandplsfordimensionreduction
AT yimingbi newmethodcombiningldaandplsfordimensionreduction
AT pengshan newmethodcombiningldaandplsfordimensionreduction
AT xiyuanhu newmethodcombiningldaandplsfordimensionreduction
_version_ 1718421954280554496