A method for speeding up feature extraction based on KPCA

Yong Xu; Dapeng Zhang; Fengxi Song; Jing Yu Yang; Zhong Jing; Miao Li

doi:10.1016/j.neucom.2006.09.005

A method for speeding up feature extraction based on KPCA

Yong Xu, Dapeng Zhang, Fengxi Song, Jing Yu Yang, Zhong Jing, Miao Li

Research output: Journal article publication › Journal article › Academic research › peer-review

134 Citations (Scopus)

Abstract

Kernel principal component analysis (KPCA) extracts features of samples with an efficiency in inverse proportion to the size of the training sample set. In this paper, we develop a novel method to improve KPCA-based feature extraction. The developed method is the first one that is methodologically consistent with KPCA. Experiments on several benchmark datasets illustrate that the feature extraction process derived from the novel method is much more efficient than that associated with KPCA. Moreover, the classification accuracy generated from the developed method is similar to that of KPCA.

Original language	English
Pages (from-to)	1056-1061
Number of pages	6
Journal	Neurocomputing
Volume	70
Issue number	4-6
DOIs	https://doi.org/10.1016/j.neucom.2006.09.005
Publication status	Published - 1 Jan 2007

Keywords

Feature extraction
Improved KPCA (IKPCA)
Kernel PCA (KPCA)
Principal component analysis (PCA)

ASJC Scopus subject areas

Computer Science Applications
Cognitive Neuroscience
Artificial Intelligence

More information

10.1016/j.neucom.2006.09.005

Cite this

@article{2d2fa657a3a145f7a9dd66f3c438ac61,

title = "A method for speeding up feature extraction based on KPCA",

abstract = "Kernel principal component analysis (KPCA) extracts features of samples with an efficiency in inverse proportion to the size of the training sample set. In this paper, we develop a novel method to improve KPCA-based feature extraction. The developed method is the first one that is methodologically consistent with KPCA. Experiments on several benchmark datasets illustrate that the feature extraction process derived from the novel method is much more efficient than that associated with KPCA. Moreover, the classification accuracy generated from the developed method is similar to that of KPCA.",

keywords = "Feature extraction, Improved KPCA (IKPCA), Kernel PCA (KPCA), Principal component analysis (PCA)",

author = "Yong Xu and Dapeng Zhang and Fengxi Song and Yang, {Jing Yu} and Zhong Jing and Miao Li",

year = "2007",

month = jan,

day = "1",

doi = "10.1016/j.neucom.2006.09.005",

language = "English",

volume = "70",

pages = "1056--1061",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier B.V.",

number = "4-6",

}

TY - JOUR

T1 - A method for speeding up feature extraction based on KPCA

AU - Xu, Yong

AU - Zhang, Dapeng

AU - Song, Fengxi

AU - Yang, Jing Yu

AU - Jing, Zhong

AU - Li, Miao

PY - 2007/1/1

Y1 - 2007/1/1

N2 - Kernel principal component analysis (KPCA) extracts features of samples with an efficiency in inverse proportion to the size of the training sample set. In this paper, we develop a novel method to improve KPCA-based feature extraction. The developed method is the first one that is methodologically consistent with KPCA. Experiments on several benchmark datasets illustrate that the feature extraction process derived from the novel method is much more efficient than that associated with KPCA. Moreover, the classification accuracy generated from the developed method is similar to that of KPCA.

AB - Kernel principal component analysis (KPCA) extracts features of samples with an efficiency in inverse proportion to the size of the training sample set. In this paper, we develop a novel method to improve KPCA-based feature extraction. The developed method is the first one that is methodologically consistent with KPCA. Experiments on several benchmark datasets illustrate that the feature extraction process derived from the novel method is much more efficient than that associated with KPCA. Moreover, the classification accuracy generated from the developed method is similar to that of KPCA.

KW - Feature extraction

KW - Improved KPCA (IKPCA)

KW - Kernel PCA (KPCA)

KW - Principal component analysis (PCA)

UR - http://www.scopus.com/inward/record.url?scp=33845973353&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2006.09.005

DO - 10.1016/j.neucom.2006.09.005

M3 - Journal article

SN - 0925-2312

VL - 70

SP - 1056

EP - 1061

JO - Neurocomputing

JF - Neurocomputing

IS - 4-6

ER -

A method for speeding up feature extraction based on KPCA

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this