Discriminant Analysis Based on Nearest Feature Line

Lijun Yan; Cong Wang; Shu-Chuan Chu; Jeng-Shyang Pan

doi:10.1007/978-3-642-36669-7_44

Discriminant Analysis Based on Nearest Feature Line

Lijun Yan, Cong Wang, Shu-Chuan Chu, Jeng-Shyang Pan

Research output: Contribution to conference › Paper › peer-review

9 Citations (Scopus)

Abstract

A novel feature extraction algorithm based on nearest feature line is proposed in this paper. The proposed algorithm can extract the local discriminant features of the samples. The performance of the proposed algorithm is directly associated with the parameter, so we use two discriminant power criterions to adaptively determine the parameter. Some experiments are implemented to evaluate the proposed algorithm and the experimental results demonstrate the efficiency of the proposed algorithm.

Original language	English
Pages	356-363
Number of pages	8
DOIs	https://doi.org/10.1007/978-3-642-36669-7_44
Publication status	Published - 1 Dec 2013
Event	Third Sino-foreign-interchange Workshop on Intelligence Science and Intelligent Data Engineering (IScIDE) - Duration: 15 Oct 2012 → …

Conference

Conference	Third Sino-foreign-interchange Workshop on Intelligence Science and Intelligent Data Engineering (IScIDE)
Period	15/10/12 → …

Keywords

Dimensionality reduction
Feature extraction
Nearest feature line

Access to Document

10.1007/978-3-642-36669-7_44

Cite this

@conference{51657d8f3f0342e5ae7c7c75c7785cc0,

title = "Discriminant Analysis Based on Nearest Feature Line",

abstract = "A novel feature extraction algorithm based on nearest feature line is proposed in this paper. The proposed algorithm can extract the local discriminant features of the samples. The performance of the proposed algorithm is directly associated with the parameter, so we use two discriminant power criterions to adaptively determine the parameter. Some experiments are implemented to evaluate the proposed algorithm and the experimental results demonstrate the efficiency of the proposed algorithm.",

keywords = "Dimensionality reduction, Feature extraction, Nearest feature line",

author = "Lijun Yan and Cong Wang and Shu-Chuan Chu and Jeng-Shyang Pan",

year = "2013",

month = dec,

day = "1",

doi = "10.1007/978-3-642-36669-7_44",

language = "English",

pages = "356--363",

note = "Third Sino-foreign-interchange Workshop on Intelligence Science and Intelligent Data Engineering (IScIDE) ; Conference date: 15-10-2012",

}

TY - CONF

T1 - Discriminant Analysis Based on Nearest Feature Line

AU - Yan, Lijun

AU - Wang, Cong

AU - Chu, Shu-Chuan

AU - Pan, Jeng-Shyang

PY - 2013/12/1

Y1 - 2013/12/1

N2 - A novel feature extraction algorithm based on nearest feature line is proposed in this paper. The proposed algorithm can extract the local discriminant features of the samples. The performance of the proposed algorithm is directly associated with the parameter, so we use two discriminant power criterions to adaptively determine the parameter. Some experiments are implemented to evaluate the proposed algorithm and the experimental results demonstrate the efficiency of the proposed algorithm.

AB - A novel feature extraction algorithm based on nearest feature line is proposed in this paper. The proposed algorithm can extract the local discriminant features of the samples. The performance of the proposed algorithm is directly associated with the parameter, so we use two discriminant power criterions to adaptively determine the parameter. Some experiments are implemented to evaluate the proposed algorithm and the experimental results demonstrate the efficiency of the proposed algorithm.

KW - Dimensionality reduction

KW - Feature extraction

KW - Nearest feature line

UR - http://www.scopus.com/inward/record.url?scp=84892926764&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-36669-7_44

DO - 10.1007/978-3-642-36669-7_44

M3 - Paper

SP - 356

EP - 363

T2 - Third Sino-foreign-interchange Workshop on Intelligence Science and Intelligent Data Engineering (IScIDE)

Y2 - 15 October 2012

ER -

Discriminant Analysis Based on Nearest Feature Line

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this