Optimization of neural network model using modified bat-inspired algorithm

Najmeh Sadat Jaddi, Salwani Abdullah, Abdul Razak Hamdan

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

The success of an artificial neural network (ANN) strongly depends on the variety of the connection weights and the network structure. Among many methods used in the literature to accurately select the network weights or structure in isolate; a few researchers have attempted to select both the weights and structure of ANN automatically by using metaheuristic algorithms. This paper proposes modified bat algorithm with a new solution representation for both optimizing the weights and structure of ANNs. The algorithm, which is based on the echolocation behaviour of bats, combines the advantages of population-based and local search algorithms. In this work, ability of the basic bat algorithm and some modified versions which are based on the consideration of the personal best solution in the velocity adjustment, the mean of personal best and global best solutions through velocity adjustment and the employment of three chaotic maps are investigated. These modifications are aimed to improve the exploration and exploitation capability of bat algorithm. Different versions of the proposed bat algorithm are incorporated to handle the selection of the structure as well as weights and biases of the ANN during the training process. We then use the Taguchi method to tune the parameters of the algorithm that demonstrates the best ability compared to the other versions. Six classifications and two time series benchmark datasets are used to test the performance of the proposed approach in terms of classification and prediction accuracy. Statistical tests demonstrate that the proposed method generates some of the best results in comparison with the latest methods in the literature. Finally, our best method is applied to a real-world problem, namely to predict the future values of rainfall data and the results show satisfactory of the method.

Original languageEnglish
Pages (from-to)71-86
Number of pages16
JournalApplied Soft Computing Journal
Volume37
DOIs
Publication statusPublished - 29 Dec 2015

Fingerprint

Neural networks
Taguchi methods
Statistical tests
Rain
Time series

Keywords

  • Artificial neural network
  • Bat-inspired algorithm
  • Chaotic map
  • Classification
  • Real-world rainfall data
  • Time series prediction

ASJC Scopus subject areas

  • Software

Cite this

Optimization of neural network model using modified bat-inspired algorithm. / Jaddi, Najmeh Sadat; Abdullah, Salwani; Hamdan, Abdul Razak.

In: Applied Soft Computing Journal, Vol. 37, 29.12.2015, p. 71-86.

Research output: Contribution to journalArticle

@article{e308e08970e242699259b7863395f78c,
title = "Optimization of neural network model using modified bat-inspired algorithm",
abstract = "The success of an artificial neural network (ANN) strongly depends on the variety of the connection weights and the network structure. Among many methods used in the literature to accurately select the network weights or structure in isolate; a few researchers have attempted to select both the weights and structure of ANN automatically by using metaheuristic algorithms. This paper proposes modified bat algorithm with a new solution representation for both optimizing the weights and structure of ANNs. The algorithm, which is based on the echolocation behaviour of bats, combines the advantages of population-based and local search algorithms. In this work, ability of the basic bat algorithm and some modified versions which are based on the consideration of the personal best solution in the velocity adjustment, the mean of personal best and global best solutions through velocity adjustment and the employment of three chaotic maps are investigated. These modifications are aimed to improve the exploration and exploitation capability of bat algorithm. Different versions of the proposed bat algorithm are incorporated to handle the selection of the structure as well as weights and biases of the ANN during the training process. We then use the Taguchi method to tune the parameters of the algorithm that demonstrates the best ability compared to the other versions. Six classifications and two time series benchmark datasets are used to test the performance of the proposed approach in terms of classification and prediction accuracy. Statistical tests demonstrate that the proposed method generates some of the best results in comparison with the latest methods in the literature. Finally, our best method is applied to a real-world problem, namely to predict the future values of rainfall data and the results show satisfactory of the method.",
keywords = "Artificial neural network, Bat-inspired algorithm, Chaotic map, Classification, Real-world rainfall data, Time series prediction",
author = "Jaddi, {Najmeh Sadat} and Salwani Abdullah and Hamdan, {Abdul Razak}",
year = "2015",
month = "12",
day = "29",
doi = "10.1016/j.asoc.2015.08.002",
language = "English",
volume = "37",
pages = "71--86",
journal = "Applied Soft Computing",
issn = "1568-4946",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Optimization of neural network model using modified bat-inspired algorithm

AU - Jaddi, Najmeh Sadat

AU - Abdullah, Salwani

AU - Hamdan, Abdul Razak

PY - 2015/12/29

Y1 - 2015/12/29

N2 - The success of an artificial neural network (ANN) strongly depends on the variety of the connection weights and the network structure. Among many methods used in the literature to accurately select the network weights or structure in isolate; a few researchers have attempted to select both the weights and structure of ANN automatically by using metaheuristic algorithms. This paper proposes modified bat algorithm with a new solution representation for both optimizing the weights and structure of ANNs. The algorithm, which is based on the echolocation behaviour of bats, combines the advantages of population-based and local search algorithms. In this work, ability of the basic bat algorithm and some modified versions which are based on the consideration of the personal best solution in the velocity adjustment, the mean of personal best and global best solutions through velocity adjustment and the employment of three chaotic maps are investigated. These modifications are aimed to improve the exploration and exploitation capability of bat algorithm. Different versions of the proposed bat algorithm are incorporated to handle the selection of the structure as well as weights and biases of the ANN during the training process. We then use the Taguchi method to tune the parameters of the algorithm that demonstrates the best ability compared to the other versions. Six classifications and two time series benchmark datasets are used to test the performance of the proposed approach in terms of classification and prediction accuracy. Statistical tests demonstrate that the proposed method generates some of the best results in comparison with the latest methods in the literature. Finally, our best method is applied to a real-world problem, namely to predict the future values of rainfall data and the results show satisfactory of the method.

AB - The success of an artificial neural network (ANN) strongly depends on the variety of the connection weights and the network structure. Among many methods used in the literature to accurately select the network weights or structure in isolate; a few researchers have attempted to select both the weights and structure of ANN automatically by using metaheuristic algorithms. This paper proposes modified bat algorithm with a new solution representation for both optimizing the weights and structure of ANNs. The algorithm, which is based on the echolocation behaviour of bats, combines the advantages of population-based and local search algorithms. In this work, ability of the basic bat algorithm and some modified versions which are based on the consideration of the personal best solution in the velocity adjustment, the mean of personal best and global best solutions through velocity adjustment and the employment of three chaotic maps are investigated. These modifications are aimed to improve the exploration and exploitation capability of bat algorithm. Different versions of the proposed bat algorithm are incorporated to handle the selection of the structure as well as weights and biases of the ANN during the training process. We then use the Taguchi method to tune the parameters of the algorithm that demonstrates the best ability compared to the other versions. Six classifications and two time series benchmark datasets are used to test the performance of the proposed approach in terms of classification and prediction accuracy. Statistical tests demonstrate that the proposed method generates some of the best results in comparison with the latest methods in the literature. Finally, our best method is applied to a real-world problem, namely to predict the future values of rainfall data and the results show satisfactory of the method.

KW - Artificial neural network

KW - Bat-inspired algorithm

KW - Chaotic map

KW - Classification

KW - Real-world rainfall data

KW - Time series prediction

UR - http://www.scopus.com/inward/record.url?scp=84940382563&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940382563&partnerID=8YFLogxK

U2 - 10.1016/j.asoc.2015.08.002

DO - 10.1016/j.asoc.2015.08.002

M3 - Article

AN - SCOPUS:84940382563

VL - 37

SP - 71

EP - 86

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

ER -