A task decomposition algorithm using radial basis functions for classification problems

Seiji Ishihara, Harukazu Igarashi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by RBF network, whose parameters are estimated according to the evaluation based on MDL criterion. In this paper, the algorithm applied for constructing a modular neural network. Experimental results showed that the algorithm simplifies multi-class classification problems effectively.

Original languageEnglish
Title of host publicationProceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005
Pages2-7
Number of pages6
Volume2005
DOIs
Publication statusPublished - 2005
EventDigital Imaging Computing: Techniques and Applications, DICTA 2005 - Cairns
Duration: 2005 Dec 62005 Dec 8

Other

OtherDigital Imaging Computing: Techniques and Applications, DICTA 2005
CityCairns
Period05/12/605/12/8

Fingerprint

Decomposition
Radial basis function networks
Neural networks

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Ishihara, S., & Igarashi, H. (2005). A task decomposition algorithm using radial basis functions for classification problems. In Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005 (Vol. 2005, pp. 2-7). [1578100] https://doi.org/10.1109/DICTA.2005.1578100

A task decomposition algorithm using radial basis functions for classification problems. / Ishihara, Seiji; Igarashi, Harukazu.

Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. Vol. 2005 2005. p. 2-7 1578100.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ishihara, S & Igarashi, H 2005, A task decomposition algorithm using radial basis functions for classification problems. in Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. vol. 2005, 1578100, pp. 2-7, Digital Imaging Computing: Techniques and Applications, DICTA 2005, Cairns, 05/12/6. https://doi.org/10.1109/DICTA.2005.1578100
Ishihara S, Igarashi H. A task decomposition algorithm using radial basis functions for classification problems. In Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. Vol. 2005. 2005. p. 2-7. 1578100 https://doi.org/10.1109/DICTA.2005.1578100
Ishihara, Seiji ; Igarashi, Harukazu. / A task decomposition algorithm using radial basis functions for classification problems. Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005. Vol. 2005 2005. pp. 2-7
@inproceedings{8a197ee493db4daf893e4f73e31c98a9,
title = "A task decomposition algorithm using radial basis functions for classification problems",
abstract = "This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by RBF network, whose parameters are estimated according to the evaluation based on MDL criterion. In this paper, the algorithm applied for constructing a modular neural network. Experimental results showed that the algorithm simplifies multi-class classification problems effectively.",
author = "Seiji Ishihara and Harukazu Igarashi",
year = "2005",
doi = "10.1109/DICTA.2005.1578100",
language = "English",
isbn = "0769524672",
volume = "2005",
pages = "2--7",
booktitle = "Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005",

}

TY - GEN

T1 - A task decomposition algorithm using radial basis functions for classification problems

AU - Ishihara, Seiji

AU - Igarashi, Harukazu

PY - 2005

Y1 - 2005

N2 - This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by RBF network, whose parameters are estimated according to the evaluation based on MDL criterion. In this paper, the algorithm applied for constructing a modular neural network. Experimental results showed that the algorithm simplifies multi-class classification problems effectively.

AB - This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by RBF network, whose parameters are estimated according to the evaluation based on MDL criterion. In this paper, the algorithm applied for constructing a modular neural network. Experimental results showed that the algorithm simplifies multi-class classification problems effectively.

UR - http://www.scopus.com/inward/record.url?scp=33846990106&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846990106&partnerID=8YFLogxK

U2 - 10.1109/DICTA.2005.1578100

DO - 10.1109/DICTA.2005.1578100

M3 - Conference contribution

SN - 0769524672

SN - 9780769524672

VL - 2005

SP - 2

EP - 7

BT - Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005

ER -