Time Series Prediction using Backpropagation Network Optimized by Hybrid K-means-Greedy Algorithm

A multilayer perceptron with backpropagation algorithm (BP) network that has the optimal number of neurons in its hidden layer would be able to predict accurately unknown values of a time series that it is trained with. A model known as K-means-Greedy Algorithm (KGA) model which combines greedy algo...

Full description

Saved in:
Bibliographic Details
Published in:Engineering Letters 2012-09, Vol.20 (3), p.203-210
Main Authors: Tan, J Y B, Bong, D B L, Rigit, A R H
Format: Article
Language:English
Subjects:
Algorithms
Back propagation
Greedy algorithms
Mathematical models
Multilayer perceptrons
Networks
Neurons
Optimization
Time series
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A multilayer perceptron with backpropagation algorithm (BP) network that has the optimal number of neurons in its hidden layer would be able to predict accurately unknown values of a time series that it is trained with. A model known as K-means-Greedy Algorithm (KGA) model which combines greedy algorithm with k-means++ clustering is proposed in this paper to find the optimal number of neurons inside the hidden layer of the BP network. Experiments performed show that the proposed KGA model is effective in finding the optimal number of neurons for the hidden layer of a BP network that is used to perform prediction of unknown values of the Mackey-Glass time series.
ISSN:1816-093X