Fuzzy aggregation of modular neural networks with ordered weighted averaging operators

Research output: Contribution to journalArticle

71 Citations (Scopus)

Abstract

This paper presents an efficient fuzzy neural system which consists of modular neural networks combined by the fuzzy integral with ordered weighted averaging (OWA) operators. The ability of the fuzzy integral to combine the results of multiple sources of information has been established in several previous works. The key point of this paper is to formalize modular neural networks as information sources, and show the feasibility of the fuzzy integral extended by OWA operators in the problem of combining neural outputs, especially in the case that the networks differ substantially from each other in accuracy. The experimental results with the recognition problem for on-line handwritten characters show that the performance of individual networks is improved significantly.

Original languageEnglish
Pages (from-to)359-375
Number of pages17
JournalInternational Journal of Approximate Reasoning
Volume13
Issue number4
DOIs
Publication statusPublished - 1995 Nov

Fingerprint

Modular Neural Networks
Fuzzy Integral
Averaging Operators
Aggregation
Agglomeration
Neural networks
Output
Experimental Results

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Artificial Intelligence
  • Applied Mathematics

Cite this

@article{a2d43b7f21de489795fbf74e1b1c8431,
title = "Fuzzy aggregation of modular neural networks with ordered weighted averaging operators",
abstract = "This paper presents an efficient fuzzy neural system which consists of modular neural networks combined by the fuzzy integral with ordered weighted averaging (OWA) operators. The ability of the fuzzy integral to combine the results of multiple sources of information has been established in several previous works. The key point of this paper is to formalize modular neural networks as information sources, and show the feasibility of the fuzzy integral extended by OWA operators in the problem of combining neural outputs, especially in the case that the networks differ substantially from each other in accuracy. The experimental results with the recognition problem for on-line handwritten characters show that the performance of individual networks is improved significantly.",
author = "{Sung-Bae Cho}, Cho",
year = "1995",
month = "11",
doi = "10.1016/0888-613X(95)00059-P",
language = "English",
volume = "13",
pages = "359--375",
journal = "International Journal of Approximate Reasoning",
issn = "0888-613X",
publisher = "Elsevier Inc.",
number = "4",

}

TY - JOUR

T1 - Fuzzy aggregation of modular neural networks with ordered weighted averaging operators

AU - Sung-Bae Cho, Cho

PY - 1995/11

Y1 - 1995/11

N2 - This paper presents an efficient fuzzy neural system which consists of modular neural networks combined by the fuzzy integral with ordered weighted averaging (OWA) operators. The ability of the fuzzy integral to combine the results of multiple sources of information has been established in several previous works. The key point of this paper is to formalize modular neural networks as information sources, and show the feasibility of the fuzzy integral extended by OWA operators in the problem of combining neural outputs, especially in the case that the networks differ substantially from each other in accuracy. The experimental results with the recognition problem for on-line handwritten characters show that the performance of individual networks is improved significantly.

AB - This paper presents an efficient fuzzy neural system which consists of modular neural networks combined by the fuzzy integral with ordered weighted averaging (OWA) operators. The ability of the fuzzy integral to combine the results of multiple sources of information has been established in several previous works. The key point of this paper is to formalize modular neural networks as information sources, and show the feasibility of the fuzzy integral extended by OWA operators in the problem of combining neural outputs, especially in the case that the networks differ substantially from each other in accuracy. The experimental results with the recognition problem for on-line handwritten characters show that the performance of individual networks is improved significantly.

UR - http://www.scopus.com/inward/record.url?scp=0000256978&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000256978&partnerID=8YFLogxK

U2 - 10.1016/0888-613X(95)00059-P

DO - 10.1016/0888-613X(95)00059-P

M3 - Article

AN - SCOPUS:0000256978

VL - 13

SP - 359

EP - 375

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

SN - 0888-613X

IS - 4

ER -