Nonlinear multiclass discriminant analysis

被引：7

作者：

Ma, JS

Sancho-Gómez, JL

Ahalt, SC

机构：

[1] Ohio State Univ, Dept Elect Engn, Columbus, OH 43210 USA

[2] Univ Carlos III Madrid, Dept Signal Theory & Commun, Getafe 28911, Spain

来源：

IEEE SIGNAL PROCESSING LETTERS | 2003年 / 10卷 / 07期

关键词：

discriminant analysis; feature extraction; kernel method;

D O I：

10.1109/LSP.2003.813680

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

An alternative nonlinear multiclass discriminant algorithm is presented. This algorithm is based on the use of kernel functions and is designed to optimize a general linear discriminant analysis criterion based on scatter matrices. By reformulating these matrices in a specific form, a straightforward derivation allows the kernel function to be introduced in a simple and direct way. Moreover, we propose a method to determine the value of the regularization parameter tau, based on this derivation.

引用

页码：196 / 199

页数：4

共 11 条

[1] Aizerman M., 1964, AUTOMAT REM CONTR, V25, P821, DOI DOI 10.1234/12345678
[2] Generalized discriminant analysis using a kernel approach
Baudat, G
Anouar, FE
[J]. NEURAL COMPUTATION, 2000, 12 (10) : 2385 - 2404
[3] Bishop C. M., 1995, NEURAL NETWORKS PATT
[4] A tutorial on Support Vector Machines for pattern recognition
Burges, CJC
[J]. DATA MINING AND KNOWLEDGE DISCOVERY, 1998, 2 (02) : 121 - 167
[5] The use of multiple measurements in taxonomic problems
Fisher, RA
[J]. ANNALS OF EUGENICS, 1936, 7 : 179 - 188
[6] Fukunaga K., 1990, INTRO STAT PATTERN R
[7] MA J, 2002, 2 REALIZATIONS GEN F
[8] Mika S., 1999, Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468), P41, DOI 10.1109/NNSP.1999.788121
[9] Roth V, 2000, ADV NEUR IN, V12, P568
[10] Nonlinear kernel-based statistical pattern analysis
Ruiz, A
López-de-Teruel, PE
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2001, 12 (01): : 16 - 32

← 1 2 →