An empirical study of design and testing of hybrid evolutionary-neural approach for classification

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).

Original languageEnglish (US)
Pages (from-to)361-374
Number of pages14
JournalOmega
Volume29
Issue number4
DOIs
StatePublished - Aug 1 2001

Fingerprint

Back propagation
Artificial neural network
Genetic algorithm
Testing
Empirical study
Evolutionary
Hybrid approach
Prediction accuracy
Overfitting
Gradient

All Science Journal Classification (ASJC) codes

  • Strategy and Management
  • Management Science and Operations Research
  • Information Systems and Management

Cite this

@article{fd99750fefcd4f20a0fdc231b9370ea4,
title = "An empirical study of design and testing of hybrid evolutionary-neural approach for classification",
abstract = "We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).",
author = "Pendharkar, {Parag C.}",
year = "2001",
month = "8",
day = "1",
doi = "10.1016/S0305-0483(01)00031-7",
language = "English (US)",
volume = "29",
pages = "361--374",
journal = "Omega",
issn = "0305-0483",
publisher = "Elsevier BV",
number = "4",

}

An empirical study of design and testing of hybrid evolutionary-neural approach for classification. / Pendharkar, Parag C.

In: Omega, Vol. 29, No. 4, 01.08.2001, p. 361-374.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An empirical study of design and testing of hybrid evolutionary-neural approach for classification

AU - Pendharkar, Parag C.

PY - 2001/8/1

Y1 - 2001/8/1

N2 - We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).

AB - We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).

UR - http://www.scopus.com/inward/record.url?scp=0038232428&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038232428&partnerID=8YFLogxK

U2 - 10.1016/S0305-0483(01)00031-7

DO - 10.1016/S0305-0483(01)00031-7

M3 - Article

AN - SCOPUS:0038232428

VL - 29

SP - 361

EP - 374

JO - Omega

JF - Omega

SN - 0305-0483

IS - 4

ER -