VARIABLE SELECTION FOR SUPPORT VECTOR MACHINES USING THE GBIC OF THE KERNEL LOGISTIC REGRESSION - Collections

＜journal article＞
VARIABLE SELECTION FOR SUPPORT VECTOR MACHINES USING THE GBIC OF THE KERNEL LOGISTIC REGRESSION

Creator	Creator Name Jiang, Ying 姜, 英 Affiliation Affiliation Name The Graduate School of Medicine, Kurume University 久留米大学大学院医学研究科
Creator	Creator Name Yanagawa, Takashi 柳川, 堯 Affiliation Affiliation Name The Biostatistics Center, Kurume University 久留米大学バイオ統計センター
Language	English
Publisher	Research Association of Statistical Sciences
Publisher	統計科学研究会
Date	2010-12
Source Title	Bulletin of informatics and cybernetics
Vol	42
First Page	45
Last Page	54
Publication Type	Version of Record
Access Rights	open access
Crossref DOI	https://doi.org/10.5109/25905
Related DOI	Bulletin of informatics and cybernetics \|\| 42 \|\| p45-54
Related DOI	http://bic.math.kyushu-u.ac.jp/
Related URI	Bulletin of informatics and cybernetics \|\| 42 \|\| p45-54
Related URI	http://bic.math.kyushu-u.ac.jp/
Relation	Bulletin of informatics and cybernetics \|\| 42 \|\| p45-54
Relation	http://bic.math.kyushu-u.ac.jp/
Abstract	In this paper we propose a method of variable selection for support vector machines based on the approximate relationship between a support vector machine and kernel logistic regression. First, we der...ive the generalized Bayesian information criterion (GBIC) of a kernel logistic regression. Then we select variables that minimize the GBIC, and propose to use them for a support vector machine. Finally, we apply the proposed method to identify peptides that could be related to pancreatic cancer.show more

File	FileType	Size	Views	Description
p045	pdf	120 KB	387

PISSN	0286-522X
NCID	AA10634475
Record ID	25905
Peer-Reviewed	Refereed
Subject Terms	variable selection
	support vector machines
	kernel logistic regression
	generalized Bayesian information criterion
	classification.
Type	学術雑誌論文
Created Date	2013.02.07
Modified Date	2020.10.26