A sequential approach for multi-class discriminant analysis with kernels

Author

Abdallah, Fahed ; Richard, Cédric ; Lengelle, Régis

Author_Institution

Lab. LM2S, Univ. de Technologie de Troyes, France

Volume

5

fYear

2004

fDate

17-21 May 2004

Abstract

Linear discriminant analysis (LDA) is a standard statistical tool for data analysis. Recently, a method called generalized discriminant analysis (GDA) has been developed to deal with nonlinear discriminant analysis using kernel functions. Difficulties for the GDA method can arise in the form of both computational complexity and storage requirements. We present a sequential algorithm for GDA avoiding these problems when one deals with large numbers of datapoints.

Keywords

computational complexity; data analysis; matrix algebra; pattern classification; statistical analysis; computational complexity; data analysis; datapoints; generalized discriminant analysis; gradient descent procedure; kernel functions; linear discriminant analysis; multi-class discriminant analysis; nonlinear discriminant analysis; pattern classification; scatter matrices; sequential algorithm; statistical tool; storage requirements; Classification algorithms; Computational complexity; Data analysis; Eigenvalues and eigenfunctions; Kernel; Linear discriminant analysis; Multi-layer neural network; Partitioning algorithms; Scattering; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-8484-9

Type

conf

DOI

10.1109/ICASSP.2004.1327145

Filename

1327145