58 International Journal of Applied Evolutionary Computation, 4(3), 58-74, July-September 2013
Copyright © 2013, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
ABSTRACT
The performance of Neural Networks (NN) depends on network structure, activation function and suitable
weight values. For fnding optimal weight values, freshly, computer scientists show the interest in the study of
social insect’s behavior learning algorithms. Chief among these are, Ant Colony Optimzation (ACO), Artifcial
Bee Colony (ABC) algorithm, Hybrid Ant Bee Colony (HABC) algorithm and Global Artifcial Bee Colony
Algorithm train Multilayer Perceptron (MLP). This paper investigates the new hybrid technique called Global
Artifcial Bee Colony-Levenberq-Marquardt (GABC-LM) algorithm. One of the crucial problems with the BP
algorithm is that it can sometimes yield the networks with suboptimal weights because of the presence of many
local optima in the solution space. To overcome GABC-LM algorithm used in this work to train MLP for the
boolean function classifcation task, the performance of GABC-LM is benchmarked against MLP training
with the typical LM, PSO, ABC and GABC. The experimental result shows that GABC-LM performs better
than that standard BP, ABC, PSO and GABC for the classifcation task.
Global Artifcial Bee Colony-
Levenberq-Marquardt (GABC-
LM) Algorithm for Classifcation
Habib Shah, Faculty of Computer Science and Information Technology, Universiti Tun
Hussein Onn Malaysia (UTHM), Parit Raja, Johor, Malaysia
Rozaida Ghazali, Faculty of Computer Science and Information Technology, Universiti Tun
Hussein Onn Malaysia (UTHM), Parit Raja, Johor, Malaysia
Nazri Mohd Nawi, Faculty of Computer Science and Information Technology, Universiti Tun
Hussein Onn Malaysia (UTHM), Parit Raja, Johor, Malaysia
Mustafa Mat Deris, Faculty of Computer Science and Information Technology, Universiti Tun
Hussein Onn Malaysia (UTHM), Parit Raja, Johor, Malaysia
Tutut Herawan, Department of Mathematics Education, Universitas Ahmad Dahlan,
Yogyakarta, Indonesia
Keywords: Ant Colony Optimization (ACO), Artifcial Bee Colony (ABC), Global Hybrid Ant Bee Colony,
Hybrid Ant Bee Colony (HABC) Algorithm, Swarm Intelligence
DOI: 10.4018/jaec.2013070106