Development of Kernel fisher discriminant model using the cross-entropy method

Budi Santosa; Andiek Sunarto

doi:10.1109/SoCPaR.2009.138

Development of Kernel fisher discriminant model using the cross-entropy method

Budi Santosa^*, Andiek Sunarto

^*Corresponding author for this work

Laboratory of Quantitative Modelling and Industrial Policy Analysis

Institut Teknologi Sepuluh Nopember

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, the cross-entropy (CE) method is proposed to solve non-linear discriminant analysis or Kernel Fisher discriminant (CE-KFD) analysis. CE through certain steps can find the optimal or near optimal solution with a fast rate of convergence for optimization problem. While, KFD is to solve problem of Fisher's linear discriminant in a kernel feature space F by maximizing between class variance and minimizing within class variance. Through the numerical experiments, we found that CE-KFD demonstrates the high accuracy of the results compared to the traditional methods, Fisher LDA and kernel Fisher (KFD) with eigen decomposition method.

Original language	English
Title of host publication	SoCPaR 2009 - Soft Computing and Pattern Recognition
Pages	691-694
Number of pages	4
DOIs	https://doi.org/10.1109/SoCPaR.2009.138
Publication status	Published - 2009
Event	International Conference on Soft Computing and Pattern Recognition, SoCPaR 2009 - Malacca, Malaysia Duration: 4 Dec 2009 → 7 Dec 2009

Publication series

Name	SoCPaR 2009 - Soft Computing and Pattern Recognition

Conference

Conference	International Conference on Soft Computing and Pattern Recognition, SoCPaR 2009
Country/Territory	Malaysia
City	Malacca
Period	4/12/09 → 7/12/09

Keywords

Accuracy
Cross entropy
Discriminant analysis
Eigen decomposition
Kernel method

Access to Document

10.1109/SoCPaR.2009.138

Cite this

@inproceedings{ef5f68b96513464a82899f842dfb2395,

title = "Development of Kernel fisher discriminant model using the cross-entropy method",

abstract = "In this paper, the cross-entropy (CE) method is proposed to solve non-linear discriminant analysis or Kernel Fisher discriminant (CE-KFD) analysis. CE through certain steps can find the optimal or near optimal solution with a fast rate of convergence for optimization problem. While, KFD is to solve problem of Fisher's linear discriminant in a kernel feature space F by maximizing between class variance and minimizing within class variance. Through the numerical experiments, we found that CE-KFD demonstrates the high accuracy of the results compared to the traditional methods, Fisher LDA and kernel Fisher (KFD) with eigen decomposition method.",

keywords = "Accuracy, Cross entropy, Discriminant analysis, Eigen decomposition, Kernel method",

author = "Budi Santosa and Andiek Sunarto",

year = "2009",

doi = "10.1109/SoCPaR.2009.138",

language = "English",

isbn = "9780769538792",

series = "SoCPaR 2009 - Soft Computing and Pattern Recognition",

pages = "691--694",

booktitle = "SoCPaR 2009 - Soft Computing and Pattern Recognition",

note = "International Conference on Soft Computing and Pattern Recognition, SoCPaR 2009 ; Conference date: 04-12-2009 Through 07-12-2009",

}

Santosa, B & Sunarto, A 2009, Development of Kernel fisher discriminant model using the cross-entropy method. in SoCPaR 2009 - Soft Computing and Pattern Recognition., 5370958, SoCPaR 2009 - Soft Computing and Pattern Recognition, pp. 691-694, International Conference on Soft Computing and Pattern Recognition, SoCPaR 2009, Malacca, Malaysia, 4/12/09. https://doi.org/10.1109/SoCPaR.2009.138

TY - GEN

T1 - Development of Kernel fisher discriminant model using the cross-entropy method

AU - Santosa, Budi

AU - Sunarto, Andiek

PY - 2009

Y1 - 2009

N2 - In this paper, the cross-entropy (CE) method is proposed to solve non-linear discriminant analysis or Kernel Fisher discriminant (CE-KFD) analysis. CE through certain steps can find the optimal or near optimal solution with a fast rate of convergence for optimization problem. While, KFD is to solve problem of Fisher's linear discriminant in a kernel feature space F by maximizing between class variance and minimizing within class variance. Through the numerical experiments, we found that CE-KFD demonstrates the high accuracy of the results compared to the traditional methods, Fisher LDA and kernel Fisher (KFD) with eigen decomposition method.

AB - In this paper, the cross-entropy (CE) method is proposed to solve non-linear discriminant analysis or Kernel Fisher discriminant (CE-KFD) analysis. CE through certain steps can find the optimal or near optimal solution with a fast rate of convergence for optimization problem. While, KFD is to solve problem of Fisher's linear discriminant in a kernel feature space F by maximizing between class variance and minimizing within class variance. Through the numerical experiments, we found that CE-KFD demonstrates the high accuracy of the results compared to the traditional methods, Fisher LDA and kernel Fisher (KFD) with eigen decomposition method.

KW - Accuracy

KW - Cross entropy

KW - Discriminant analysis

KW - Eigen decomposition

KW - Kernel method

UR - http://www.scopus.com/inward/record.url?scp=77649323593&partnerID=8YFLogxK

U2 - 10.1109/SoCPaR.2009.138

DO - 10.1109/SoCPaR.2009.138

M3 - Conference contribution

AN - SCOPUS:77649323593

SN - 9780769538792

T3 - SoCPaR 2009 - Soft Computing and Pattern Recognition

SP - 691

EP - 694

BT - SoCPaR 2009 - Soft Computing and Pattern Recognition

T2 - International Conference on Soft Computing and Pattern Recognition, SoCPaR 2009

Y2 - 4 December 2009 through 7 December 2009

ER -

Development of Kernel fisher discriminant model using the cross-entropy method

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this