A neuro-expert system model for conflict resolution

N. V. Subba Reddy, P. Nagabhushan

Research output: Contribution to journalArticle

Abstract

We describe here a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The neural network classifier is a combination of modified self-organizing map and learning vector quantization. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions generated by the neural network. The results obtained from this two-tier architecture are compared with those of a combination of four algorithms. This work shows that it is possible to eliminate the substitution while maintaining a fairly high recognition.

Original languageEnglish
Pages (from-to)647-650
Number of pages4
JournalCurrent Science
Volume72
Issue number9
Publication statusPublished - 10-05-1997

Fingerprint

Expert systems
Neural networks
Vector quantization
Self organizing maps
Classifiers
Substitution reactions

All Science Journal Classification (ASJC) codes

  • General

Cite this

Subba Reddy, N. V. ; Nagabhushan, P. / A neuro-expert system model for conflict resolution. In: Current Science. 1997 ; Vol. 72, No. 9. pp. 647-650.
@article{8074d40f1e424bc5914186890959b230,
title = "A neuro-expert system model for conflict resolution",
abstract = "We describe here a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The neural network classifier is a combination of modified self-organizing map and learning vector quantization. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions generated by the neural network. The results obtained from this two-tier architecture are compared with those of a combination of four algorithms. This work shows that it is possible to eliminate the substitution while maintaining a fairly high recognition.",
author = "{Subba Reddy}, {N. V.} and P. Nagabhushan",
year = "1997",
month = "5",
day = "10",
language = "English",
volume = "72",
pages = "647--650",
journal = "Current Science",
issn = "0011-3891",
publisher = "Indian Academy of Sciences",
number = "9",

}

Subba Reddy, NV & Nagabhushan, P 1997, 'A neuro-expert system model for conflict resolution', Current Science, vol. 72, no. 9, pp. 647-650.

A neuro-expert system model for conflict resolution. / Subba Reddy, N. V.; Nagabhushan, P.

In: Current Science, Vol. 72, No. 9, 10.05.1997, p. 647-650.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A neuro-expert system model for conflict resolution

AU - Subba Reddy, N. V.

AU - Nagabhushan, P.

PY - 1997/5/10

Y1 - 1997/5/10

N2 - We describe here a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The neural network classifier is a combination of modified self-organizing map and learning vector quantization. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions generated by the neural network. The results obtained from this two-tier architecture are compared with those of a combination of four algorithms. This work shows that it is possible to eliminate the substitution while maintaining a fairly high recognition.

AB - We describe here a neural network and expert system model for conflict resolution of unconstrained handwritten numerals. The neural network classifier is a combination of modified self-organizing map and learning vector quantization. The basic recognizer is the neural network. It solves most of the cases, but fails in certain confusing cases. The expert system, the second recognizer, resolves the confusions generated by the neural network. The results obtained from this two-tier architecture are compared with those of a combination of four algorithms. This work shows that it is possible to eliminate the substitution while maintaining a fairly high recognition.

UR - http://www.scopus.com/inward/record.url?scp=23744482299&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=23744482299&partnerID=8YFLogxK

M3 - Article

VL - 72

SP - 647

EP - 650

JO - Current Science

JF - Current Science

SN - 0011-3891

IS - 9

ER -