Publication:
Distance weighted K-means algorithm for center selection in training radial basis function networks

Simple item page
cris.author.scopus-author-id 23491376300
cris.author.scopus-author-id 56416988300
cris.author.scopus-author-id 38561331300
cris.virtual.department Universiti Malaysia Perlis
cris.virtualsource.department 97249ac5-6bf2-440d-b171-28ee875c9f9c
dc.contributor.author Aik L.E.
dc.contributor.author Hong T.W.
dc.contributor.author Ahmad Kadri Junoh
dc.date.accessioned 2024-12-11T07:56:06Z
dc.date.available 2024-12-11T07:56:06Z
dc.date.issued 2019-03-01
dc.description.abstract The accuracies rates of the neural networks mainly depend on the selection of the correct data centers. The K-means algorithm is a widely used clustering algorithm in various disciplines for centers selection. However, the method is known for its sensitivity to initial centers selection. It suffers not only from a high dependency on the algorithm's initial centers selection but, also from data points. The performance of K-means has been enhanced from different perspectives, including centroid initialization problem over the years. Unfortunately, the solution does not provide a good trade-off between quality and efficiency of the centers produces by the algorithm. To solve this problem, a new method to find the initial centers and improve the sensitivity to the initial centers of K-means algorithm is proposed. This paper presented a training algorithm for the radial basis function network (RBFN) using improved K-means (KM) algorithm, which is the modified version of KM algorithm based on distance-weighted adjustment for each centers, known as distance-weighted K-means (DWKM) algorithm. The proposed training algorithm, which uses DWKM algorithm select centers for training RBFN obtained better accuracy in predictions and reduced network architecture compared to the standard RBFN. The proposed training algorithm was implemented in MATLAB environment; hence, the new network was undergoing a hybrid learning process. The network called DWKM-RBFN was tested against the standard RBFN in predictions. The experimental models were tested on four literatures nonlinear function and four real-world application problems, particularly in Air pollutant problem, Biochemical Oxygen Demand (BOD) problem, Phytoplankton problem, and forex pair EURUSD. The results are compared to proposed method for root mean square error (RMSE) in radial basis function network (RBFN). The proposed method yielded a promising result with an average improvement percentage more than 50 percent in RMSE.
dc.identifier.doi 10.11591/ijai.v8.i1.pp54-62
dc.identifier.scopus 2-s2.0-85068418034
dc.identifier.uri https://hdl.handle.net/20.500.14170/10115
dc.relation.grantno undefined
dc.relation.ispartof IAES International Journal of Artificial Intelligence
dc.relation.ispartofseries IAES International Journal of Artificial Intelligence
dc.relation.issn 20894872
dc.rights open access
dc.subject Center selection | Dataset size reduction | K-means algorithm | Neural network | Radial basis function network
dc.title Distance weighted K-means algorithm for center selection in training radial basis function networks
dc.type Journal
dspace.entity.type Publication
oaire.citation.endPage 62
oaire.citation.issue 1
oaire.citation.startPage 54
oaire.citation.volume 8
oairecerif.affiliation.orgunit Universiti Malaysia Perlis
oairecerif.affiliation.orgunit Universiti Malaysia Perlis
oairecerif.affiliation.orgunit Universiti Malaysia Perlis
oairecerif.author.affiliation #PLACEHOLDER_PARENT_METADATA_VALUE#
oairecerif.author.affiliation #PLACEHOLDER_PARENT_METADATA_VALUE#
oairecerif.author.affiliation Universiti Malaysia Perlis
person.identifier.orcid #PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.orcid #PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.orcid #PLACEHOLDER_PARENT_METADATA_VALUE#
person.identifier.scopus-author-id 23491376300
person.identifier.scopus-author-id 56416988300
person.identifier.scopus-author-id 38561331300
Files
Collections
Publications 2019