Improved GART neural network model for pattern classification and rule extraction with application to power systems

Keem Siah Yap, Chee Peng Lim, Mau Teng Au

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Generalized adaptive resonance theory (GART) is a neural network model that is capable of online learning and is effective in tackling pattern classification tasks. In this paper, we propose an improved GART model (IGART), and demonstrate its applicability to power systems. IGART enhances the dynamics of GART in several aspects, which include the use of the Laplacian likelihood function, a new vigilance function, a new match-tracking mechanism, an ordering algorithm for determining the sequence of training data, and a rule extraction capability to elicit if-then rules from the network. To assess the effectiveness of IGART and to compare its performances with those from other methods, three datasets that are related to power systems are employed. The experimental results demonstrate the usefulness of IGART with the rule extraction capability in undertaking classification problems in power systems engineering.

Original languageEnglish
Article number6069866
Pages (from-to)2310-2323
Number of pages14
JournalIEEE Transactions on Neural Networks
Volume22
Issue number12 PART 2
DOIs
Publication statusPublished - 01 Dec 2011

Fingerprint

Pattern recognition
Neural networks
Systems engineering

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

@article{1b0a47be5224416ca8feefb130dd09ab,
title = "Improved GART neural network model for pattern classification and rule extraction with application to power systems",
abstract = "Generalized adaptive resonance theory (GART) is a neural network model that is capable of online learning and is effective in tackling pattern classification tasks. In this paper, we propose an improved GART model (IGART), and demonstrate its applicability to power systems. IGART enhances the dynamics of GART in several aspects, which include the use of the Laplacian likelihood function, a new vigilance function, a new match-tracking mechanism, an ordering algorithm for determining the sequence of training data, and a rule extraction capability to elicit if-then rules from the network. To assess the effectiveness of IGART and to compare its performances with those from other methods, three datasets that are related to power systems are employed. The experimental results demonstrate the usefulness of IGART with the rule extraction capability in undertaking classification problems in power systems engineering.",
author = "Yap, {Keem Siah} and Lim, {Chee Peng} and Au, {Mau Teng}",
year = "2011",
month = "12",
day = "1",
doi = "10.1109/TNN.2011.2173502",
language = "English",
volume = "22",
pages = "2310--2323",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "12 PART 2",

}

Improved GART neural network model for pattern classification and rule extraction with application to power systems. / Yap, Keem Siah; Lim, Chee Peng; Au, Mau Teng.

In: IEEE Transactions on Neural Networks, Vol. 22, No. 12 PART 2, 6069866, 01.12.2011, p. 2310-2323.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Improved GART neural network model for pattern classification and rule extraction with application to power systems

AU - Yap, Keem Siah

AU - Lim, Chee Peng

AU - Au, Mau Teng

PY - 2011/12/1

Y1 - 2011/12/1

N2 - Generalized adaptive resonance theory (GART) is a neural network model that is capable of online learning and is effective in tackling pattern classification tasks. In this paper, we propose an improved GART model (IGART), and demonstrate its applicability to power systems. IGART enhances the dynamics of GART in several aspects, which include the use of the Laplacian likelihood function, a new vigilance function, a new match-tracking mechanism, an ordering algorithm for determining the sequence of training data, and a rule extraction capability to elicit if-then rules from the network. To assess the effectiveness of IGART and to compare its performances with those from other methods, three datasets that are related to power systems are employed. The experimental results demonstrate the usefulness of IGART with the rule extraction capability in undertaking classification problems in power systems engineering.

AB - Generalized adaptive resonance theory (GART) is a neural network model that is capable of online learning and is effective in tackling pattern classification tasks. In this paper, we propose an improved GART model (IGART), and demonstrate its applicability to power systems. IGART enhances the dynamics of GART in several aspects, which include the use of the Laplacian likelihood function, a new vigilance function, a new match-tracking mechanism, an ordering algorithm for determining the sequence of training data, and a rule extraction capability to elicit if-then rules from the network. To assess the effectiveness of IGART and to compare its performances with those from other methods, three datasets that are related to power systems are employed. The experimental results demonstrate the usefulness of IGART with the rule extraction capability in undertaking classification problems in power systems engineering.

UR - http://www.scopus.com/inward/record.url?scp=83655167009&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=83655167009&partnerID=8YFLogxK

U2 - 10.1109/TNN.2011.2173502

DO - 10.1109/TNN.2011.2173502

M3 - Article

VL - 22

SP - 2310

EP - 2323

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 12 PART 2

M1 - 6069866

ER -