Taguchi-based parameter designing of genetic algorithm for artificial neural network training

Najmeh Sadat Jaddi, Salwani Abdullah, Abdul Razak Hamdan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Citations (Scopus)

Abstract

A number of properties of Artificial Neural Networks (ANNs) make them suitable for many applications such as time series prediction problem. However, lack of training model which finds a global optimal set of weights has been disadvantaged in some real-world problems. Genetic algorithm is an optimization procedure which is superior at exploring a search space in an intelligent method. In this paper we present a genetic-based algorithm to optimize the weights and biases of the ANN. In this work we tune the parameters of the genetic algorithm using Taguchi method. To test the method two standard time series prediction problems are employed. The results are compared to the methods in the literature. The comparison showed the superiority of the proposed method.

Original languageEnglish
Title of host publicationProceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013
PublisherIEEE Computer Society
Pages278-281
Number of pages4
ISBN (Print)9780769551333
DOIs
Publication statusPublished - 2013
Event2013 International Conference on Informatics and Creative Multimedia, ICICM 2013 - Kuala Lumpur
Duration: 4 Sep 20136 Sep 2013

Other

Other2013 International Conference on Informatics and Creative Multimedia, ICICM 2013
CityKuala Lumpur
Period4/9/136/9/13

Fingerprint

Time series
Genetic algorithms
Neural networks
Taguchi methods

Keywords

  • Artificial neural network training
  • Genetic algorithm
  • Taguchi method
  • Time series prediction

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Information Systems

Cite this

Jaddi, N. S., Abdullah, S., & Hamdan, A. R. (2013). Taguchi-based parameter designing of genetic algorithm for artificial neural network training. In Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013 (pp. 278-281). [6702824] IEEE Computer Society. https://doi.org/10.1109/ICICM.2013.54

Taguchi-based parameter designing of genetic algorithm for artificial neural network training. / Jaddi, Najmeh Sadat; Abdullah, Salwani; Hamdan, Abdul Razak.

Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013. IEEE Computer Society, 2013. p. 278-281 6702824.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jaddi, NS, Abdullah, S & Hamdan, AR 2013, Taguchi-based parameter designing of genetic algorithm for artificial neural network training. in Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013., 6702824, IEEE Computer Society, pp. 278-281, 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013, Kuala Lumpur, 4/9/13. https://doi.org/10.1109/ICICM.2013.54
Jaddi NS, Abdullah S, Hamdan AR. Taguchi-based parameter designing of genetic algorithm for artificial neural network training. In Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013. IEEE Computer Society. 2013. p. 278-281. 6702824 https://doi.org/10.1109/ICICM.2013.54
Jaddi, Najmeh Sadat ; Abdullah, Salwani ; Hamdan, Abdul Razak. / Taguchi-based parameter designing of genetic algorithm for artificial neural network training. Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013. IEEE Computer Society, 2013. pp. 278-281
@inproceedings{3c67e59af290485ab19d93834589d214,
title = "Taguchi-based parameter designing of genetic algorithm for artificial neural network training",
abstract = "A number of properties of Artificial Neural Networks (ANNs) make them suitable for many applications such as time series prediction problem. However, lack of training model which finds a global optimal set of weights has been disadvantaged in some real-world problems. Genetic algorithm is an optimization procedure which is superior at exploring a search space in an intelligent method. In this paper we present a genetic-based algorithm to optimize the weights and biases of the ANN. In this work we tune the parameters of the genetic algorithm using Taguchi method. To test the method two standard time series prediction problems are employed. The results are compared to the methods in the literature. The comparison showed the superiority of the proposed method.",
keywords = "Artificial neural network training, Genetic algorithm, Taguchi method, Time series prediction",
author = "Jaddi, {Najmeh Sadat} and Salwani Abdullah and Hamdan, {Abdul Razak}",
year = "2013",
doi = "10.1109/ICICM.2013.54",
language = "English",
isbn = "9780769551333",
pages = "278--281",
booktitle = "Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Taguchi-based parameter designing of genetic algorithm for artificial neural network training

AU - Jaddi, Najmeh Sadat

AU - Abdullah, Salwani

AU - Hamdan, Abdul Razak

PY - 2013

Y1 - 2013

N2 - A number of properties of Artificial Neural Networks (ANNs) make them suitable for many applications such as time series prediction problem. However, lack of training model which finds a global optimal set of weights has been disadvantaged in some real-world problems. Genetic algorithm is an optimization procedure which is superior at exploring a search space in an intelligent method. In this paper we present a genetic-based algorithm to optimize the weights and biases of the ANN. In this work we tune the parameters of the genetic algorithm using Taguchi method. To test the method two standard time series prediction problems are employed. The results are compared to the methods in the literature. The comparison showed the superiority of the proposed method.

AB - A number of properties of Artificial Neural Networks (ANNs) make them suitable for many applications such as time series prediction problem. However, lack of training model which finds a global optimal set of weights has been disadvantaged in some real-world problems. Genetic algorithm is an optimization procedure which is superior at exploring a search space in an intelligent method. In this paper we present a genetic-based algorithm to optimize the weights and biases of the ANN. In this work we tune the parameters of the genetic algorithm using Taguchi method. To test the method two standard time series prediction problems are employed. The results are compared to the methods in the literature. The comparison showed the superiority of the proposed method.

KW - Artificial neural network training

KW - Genetic algorithm

KW - Taguchi method

KW - Time series prediction

UR - http://www.scopus.com/inward/record.url?scp=84893647969&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893647969&partnerID=8YFLogxK

U2 - 10.1109/ICICM.2013.54

DO - 10.1109/ICICM.2013.54

M3 - Conference contribution

AN - SCOPUS:84893647969

SN - 9780769551333

SP - 278

EP - 281

BT - Proceedings - 2013 International Conference on Informatics and Creative Multimedia, ICICM 2013

PB - IEEE Computer Society

ER -