Autonomous and deterministic probabilistic neural network using global k-means

We present a comparative study between Expectation-Maximization (EM) trained probabilistic neural network (PNN) with random initialization and with initialization from Global k-means. To make the results more comprehensive, the algorithm was tested on both homoscedastic and heteroscedastic PNNs. Nor...

Full description

Bibliographic Details
Main Authors: Chang, Roy Kwang Yang, Loo, Chu Kiong, Chu Kiong, Rao, , Machavaram V. C.
Format: Article
Published: 2006
Subjects:
Online Access:http://shdl.mmu.edu.my/2066/
_version_ 1848789955162144768
author Chang, Roy Kwang Yang
Loo, Chu Kiong, Chu Kiong
Rao, , Machavaram V. C.
author_facet Chang, Roy Kwang Yang
Loo, Chu Kiong, Chu Kiong
Rao, , Machavaram V. C.
author_sort Chang, Roy Kwang Yang
building MMU Institutional Repository
collection Online Access
description We present a comparative study between Expectation-Maximization (EM) trained probabilistic neural network (PNN) with random initialization and with initialization from Global k-means. To make the results more comprehensive, the algorithm was tested on both homoscedastic and heteroscedastic PNNs. Normally, user have to define the number of clusters through trial and error method, which makes random initialization to be of stochastic nature. Global k-means was chosen as the initialization method because it can autonomously find the number of clusters using a selection criterion and can provide deterministic clustering results. The proposed algorithm was tested on benchmark datasets and real world data from the cooling water system in a power plant.
first_indexed 2025-11-14T18:04:56Z
format Article
id mmu-2066
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:04:56Z
publishDate 2006
recordtype eprints
repository_type Digital Repository
spelling mmu-20662011-08-10T07:04:46Z http://shdl.mmu.edu.my/2066/ Autonomous and deterministic probabilistic neural network using global k-means Chang, Roy Kwang Yang Loo, Chu Kiong, Chu Kiong Rao, , Machavaram V. C. QA75.5-76.95 Electronic computers. Computer science We present a comparative study between Expectation-Maximization (EM) trained probabilistic neural network (PNN) with random initialization and with initialization from Global k-means. To make the results more comprehensive, the algorithm was tested on both homoscedastic and heteroscedastic PNNs. Normally, user have to define the number of clusters through trial and error method, which makes random initialization to be of stochastic nature. Global k-means was chosen as the initialization method because it can autonomously find the number of clusters using a selection criterion and can provide deterministic clustering results. The proposed algorithm was tested on benchmark datasets and real world data from the cooling water system in a power plant. 2006 Article NonPeerReviewed Chang, Roy Kwang Yang and Loo, Chu Kiong, Chu Kiong and Rao, , Machavaram V. C. (2006) Autonomous and deterministic probabilistic neural network using global k-means. ADVANCES IN NEURAL NETWORKS - ISNN 2006, 3971 (1). pp. 830-836. ISSN 0302-9743
spellingShingle QA75.5-76.95 Electronic computers. Computer science
Chang, Roy Kwang Yang
Loo, Chu Kiong, Chu Kiong
Rao, , Machavaram V. C.
Autonomous and deterministic probabilistic neural network using global k-means
title Autonomous and deterministic probabilistic neural network using global k-means
title_full Autonomous and deterministic probabilistic neural network using global k-means
title_fullStr Autonomous and deterministic probabilistic neural network using global k-means
title_full_unstemmed Autonomous and deterministic probabilistic neural network using global k-means
title_short Autonomous and deterministic probabilistic neural network using global k-means
title_sort autonomous and deterministic probabilistic neural network using global k-means
topic QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/2066/