Hybridizing firefly algorithms with a probabilistic neural network for solving classification problems

Mohammed Alweshah, Salwani Abdullah

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

Classification is one of the important tasks in data mining. The probabilistic neural network (PNN) is a well-known and efficient approach for classification. The objective of the work presented in this paper is to build on this approach to develop an effective method for classification problems that can find high-quality solutions (with respect to classification accuracy) at a high convergence speed. To achieve this objective, we propose a method that hybridizes the firefly algorithm with simulated annealing (denoted as SFA), where simulated annealing is applied to control the randomness step inside the firefly algorithm while optimizing the weights of the standard PNN model. We also extend our work by investigating the effectiveness of using Lévy flight within the firefly algorithm (denoted as LFA) to better explore the search space and by integrating SFA with Lévy flight (denoted as LSFA) in order to improve the performance of the PNN. The algorithms were tested on 11 standard benchmark datasets. Experimental results indicate that the LSFA shows better performance than the SFA and LFA. Moreover, when compared with other algorithms in the literature, the LSFA is able to obtain better results in terms of classification accuracy.

Original languageEnglish
Pages (from-to)513-524
Number of pages12
JournalApplied Soft Computing Journal
Volume35
DOIs
Publication statusPublished - 20 Jul 2015

Fingerprint

Neural networks
Simulated annealing
Data mining

Keywords

  • Classification problems
  • Firefly algorithm
  • Lévy flight
  • Probabilistic neural networks
  • Simulated annealing

ASJC Scopus subject areas

  • Software

Cite this

Hybridizing firefly algorithms with a probabilistic neural network for solving classification problems. / Alweshah, Mohammed; Abdullah, Salwani.

In: Applied Soft Computing Journal, Vol. 35, 20.07.2015, p. 513-524.

Research output: Contribution to journalArticle

@article{0c3a8ae06d2240468d000ac7585f59ce,
title = "Hybridizing firefly algorithms with a probabilistic neural network for solving classification problems",
abstract = "Classification is one of the important tasks in data mining. The probabilistic neural network (PNN) is a well-known and efficient approach for classification. The objective of the work presented in this paper is to build on this approach to develop an effective method for classification problems that can find high-quality solutions (with respect to classification accuracy) at a high convergence speed. To achieve this objective, we propose a method that hybridizes the firefly algorithm with simulated annealing (denoted as SFA), where simulated annealing is applied to control the randomness step inside the firefly algorithm while optimizing the weights of the standard PNN model. We also extend our work by investigating the effectiveness of using L{\'e}vy flight within the firefly algorithm (denoted as LFA) to better explore the search space and by integrating SFA with L{\'e}vy flight (denoted as LSFA) in order to improve the performance of the PNN. The algorithms were tested on 11 standard benchmark datasets. Experimental results indicate that the LSFA shows better performance than the SFA and LFA. Moreover, when compared with other algorithms in the literature, the LSFA is able to obtain better results in terms of classification accuracy.",
keywords = "Classification problems, Firefly algorithm, L{\'e}vy flight, Probabilistic neural networks, Simulated annealing",
author = "Mohammed Alweshah and Salwani Abdullah",
year = "2015",
month = "7",
day = "20",
doi = "10.1016/j.asoc.2015.06.018",
language = "English",
volume = "35",
pages = "513--524",
journal = "Applied Soft Computing",
issn = "1568-4946",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Hybridizing firefly algorithms with a probabilistic neural network for solving classification problems

AU - Alweshah, Mohammed

AU - Abdullah, Salwani

PY - 2015/7/20

Y1 - 2015/7/20

N2 - Classification is one of the important tasks in data mining. The probabilistic neural network (PNN) is a well-known and efficient approach for classification. The objective of the work presented in this paper is to build on this approach to develop an effective method for classification problems that can find high-quality solutions (with respect to classification accuracy) at a high convergence speed. To achieve this objective, we propose a method that hybridizes the firefly algorithm with simulated annealing (denoted as SFA), where simulated annealing is applied to control the randomness step inside the firefly algorithm while optimizing the weights of the standard PNN model. We also extend our work by investigating the effectiveness of using Lévy flight within the firefly algorithm (denoted as LFA) to better explore the search space and by integrating SFA with Lévy flight (denoted as LSFA) in order to improve the performance of the PNN. The algorithms were tested on 11 standard benchmark datasets. Experimental results indicate that the LSFA shows better performance than the SFA and LFA. Moreover, when compared with other algorithms in the literature, the LSFA is able to obtain better results in terms of classification accuracy.

AB - Classification is one of the important tasks in data mining. The probabilistic neural network (PNN) is a well-known and efficient approach for classification. The objective of the work presented in this paper is to build on this approach to develop an effective method for classification problems that can find high-quality solutions (with respect to classification accuracy) at a high convergence speed. To achieve this objective, we propose a method that hybridizes the firefly algorithm with simulated annealing (denoted as SFA), where simulated annealing is applied to control the randomness step inside the firefly algorithm while optimizing the weights of the standard PNN model. We also extend our work by investigating the effectiveness of using Lévy flight within the firefly algorithm (denoted as LFA) to better explore the search space and by integrating SFA with Lévy flight (denoted as LSFA) in order to improve the performance of the PNN. The algorithms were tested on 11 standard benchmark datasets. Experimental results indicate that the LSFA shows better performance than the SFA and LFA. Moreover, when compared with other algorithms in the literature, the LSFA is able to obtain better results in terms of classification accuracy.

KW - Classification problems

KW - Firefly algorithm

KW - Lévy flight

KW - Probabilistic neural networks

KW - Simulated annealing

UR - http://www.scopus.com/inward/record.url?scp=84937559609&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937559609&partnerID=8YFLogxK

U2 - 10.1016/j.asoc.2015.06.018

DO - 10.1016/j.asoc.2015.06.018

M3 - Article

AN - SCOPUS:84937559609

VL - 35

SP - 513

EP - 524

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

ER -