A Task Decomposition Algorithm Based on the Distribution of Input Pattern Vectors for Classification Problems

Seiji Ishihara, Harukazu Igarashi

Research output: Contribution to journalArticle

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 corresponding to each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by Gaussian mixture models which are estimated by EM algorithm with 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 efficiently.

Original languageEnglish
Pages (from-to)1043-1048
Number of pages6
JournalIEEJ Transactions on Electronics, Information and Systems
Volume125
Issue number7
DOIs
Publication statusPublished - 2005

Fingerprint

Decomposition
Neural networks

Keywords

  • classification problem
  • EM algorithm
  • minimum description length criterion
  • modular neural network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

@article{5d3389e7daf24daa8b70ab0782c9bff6,
title = "A Task Decomposition Algorithm Based on the Distribution of Input Pattern Vectors 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 corresponding to each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by Gaussian mixture models which are estimated by EM algorithm with 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 efficiently.",
keywords = "classification problem, EM algorithm, minimum description length criterion, modular neural network",
author = "Seiji Ishihara and Harukazu Igarashi",
year = "2005",
doi = "10.1541/ieejeiss.125.1043",
language = "English",
volume = "125",
pages = "1043--1048",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "7",

}

TY - JOUR

T1 - A Task Decomposition Algorithm Based on the Distribution of Input Pattern Vectors 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 corresponding to each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by Gaussian mixture models which are estimated by EM algorithm with 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 efficiently.

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 corresponding to each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by Gaussian mixture models which are estimated by EM algorithm with 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 efficiently.

KW - classification problem

KW - EM algorithm

KW - minimum description length criterion

KW - modular neural network

UR - http://www.scopus.com/inward/record.url?scp=85024721631&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024721631&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.125.1043

DO - 10.1541/ieejeiss.125.1043

M3 - Article

AN - SCOPUS:85024721631

VL - 125

SP - 1043

EP - 1048

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 7

ER -