RADIAL BASIS FUNCTION NEURAL NETWORKS

Loading...
Thumbnail Image
Date
2005-05
Authors
WOI, CALEN
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A training algorithm for training radial basis function neural networks is presented in this dissertation. The training algorithm used to train the network is based on the subtractive clustering method which has a number of advantages such as faster training time and reduced network architecture compared to the standard radial basis function. The proposed training algorithm was implemented in the radial basis neural network created by the newrb function from MA TLAB which already uses gradient based iterative method as the learning strategy, therefore the new network will undergo a hybrid learning process. The network, called SC/RBF (Subtractive Clustering - Radial Basis Neural Network) was tested against the standard radial basis neural network in function approximations and forecasting. Since the use of neural network techniques to predict air pollutant trend is considered as a reliable yet cost-effective method for the purpose, the experimental model was tested by forecasting the pollutant trend at Forth Worth City, Texas with air quality data from Texas Natural Resources Conservation Commission database.
Description
Keywords
RADIAL BASIS , NEURAL NETWORKS
Citation