A classification approach using multi-layered neural networks

Selwyn Piramuthu, Michael J. Shaw, James A. Gentry

Research output: Contribution to journalArticle

Abstract

There has been an increasing interest in the applicability of neural networks in disparate domains. In this paper, we describe the use of multi-layered perceptrons, a type of neural-network topology, for financial classification problems, with promising results. Back-propagation, which is the learning algorithm most often used in multi-layered perceptrons, however, is inherently an inefficient search procedure. We present improved procedures which have much better convergence properties. Using several financial classification applications as examples, we show the efficacy of using multilayered perceptrons with improved learning algorithms. The modified learning algorithms have better performance, in terms of classification/prediction accuracies, than the methods previously used in the literature, such as probit analysis and similarity-based learning techniques.

Original languageEnglish (US)
Pages (from-to)509-525
Number of pages17
JournalDecision Support Systems
Volume11
Issue number5
DOIs
StatePublished - Jun 1994

Fingerprint

Neural Networks (Computer)
Learning
Neural networks
Learning algorithms
Backpropagation
Topology
Learning algorithm
Neural Networks
Network topology
Probit analysis
Back propagation
Efficacy
Prediction accuracy

Keywords

  • Back-propagation
  • Classification
  • Gradient-search
  • Neural networks

ASJC Scopus subject areas

  • Management Information Systems
  • Information Systems
  • Developmental and Educational Psychology
  • Arts and Humanities (miscellaneous)
  • Information Systems and Management

Cite this

A classification approach using multi-layered neural networks. / Piramuthu, Selwyn; Shaw, Michael J.; Gentry, James A.

In: Decision Support Systems, Vol. 11, No. 5, 06.1994, p. 509-525.

Research output: Contribution to journalArticle

Piramuthu, Selwyn ; Shaw, Michael J. ; Gentry, James A. / A classification approach using multi-layered neural networks. In: Decision Support Systems. 1994 ; Vol. 11, No. 5. pp. 509-525.
@article{3884347ed4a94316a60ed3a26e0eff6b,
title = "A classification approach using multi-layered neural networks",
abstract = "There has been an increasing interest in the applicability of neural networks in disparate domains. In this paper, we describe the use of multi-layered perceptrons, a type of neural-network topology, for financial classification problems, with promising results. Back-propagation, which is the learning algorithm most often used in multi-layered perceptrons, however, is inherently an inefficient search procedure. We present improved procedures which have much better convergence properties. Using several financial classification applications as examples, we show the efficacy of using multilayered perceptrons with improved learning algorithms. The modified learning algorithms have better performance, in terms of classification/prediction accuracies, than the methods previously used in the literature, such as probit analysis and similarity-based learning techniques.",
keywords = "Back-propagation, Classification, Gradient-search, Neural networks",
author = "Selwyn Piramuthu and Shaw, {Michael J.} and Gentry, {James A.}",
year = "1994",
month = "6",
doi = "10.1016/0167-9236(94)90022-1",
language = "English (US)",
volume = "11",
pages = "509--525",
journal = "Decision Support Systems",
issn = "0167-9236",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - A classification approach using multi-layered neural networks

AU - Piramuthu, Selwyn

AU - Shaw, Michael J.

AU - Gentry, James A.

PY - 1994/6

Y1 - 1994/6

N2 - There has been an increasing interest in the applicability of neural networks in disparate domains. In this paper, we describe the use of multi-layered perceptrons, a type of neural-network topology, for financial classification problems, with promising results. Back-propagation, which is the learning algorithm most often used in multi-layered perceptrons, however, is inherently an inefficient search procedure. We present improved procedures which have much better convergence properties. Using several financial classification applications as examples, we show the efficacy of using multilayered perceptrons with improved learning algorithms. The modified learning algorithms have better performance, in terms of classification/prediction accuracies, than the methods previously used in the literature, such as probit analysis and similarity-based learning techniques.

AB - There has been an increasing interest in the applicability of neural networks in disparate domains. In this paper, we describe the use of multi-layered perceptrons, a type of neural-network topology, for financial classification problems, with promising results. Back-propagation, which is the learning algorithm most often used in multi-layered perceptrons, however, is inherently an inefficient search procedure. We present improved procedures which have much better convergence properties. Using several financial classification applications as examples, we show the efficacy of using multilayered perceptrons with improved learning algorithms. The modified learning algorithms have better performance, in terms of classification/prediction accuracies, than the methods previously used in the literature, such as probit analysis and similarity-based learning techniques.

KW - Back-propagation

KW - Classification

KW - Gradient-search

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=0028448573&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028448573&partnerID=8YFLogxK

U2 - 10.1016/0167-9236(94)90022-1

DO - 10.1016/0167-9236(94)90022-1

M3 - Article

AN - SCOPUS:0028448573

VL - 11

SP - 509

EP - 525

JO - Decision Support Systems

JF - Decision Support Systems

SN - 0167-9236

IS - 5

ER -