WSEAS CONFERENCES. WSEAS, Unifying the Science

Main Page of the Journal                                                           Main Page of the WSEAS

October 2007, November 2007, December 2007, 2008


Issue 12, Volume 6, December 2007
Print ISSN: 1109-2769
E-ISSN: 2224-2880








Title of the Paper:  The On-line Cross Entropy Method for Unsupervised Data Exploration


Authors: Ying Wu, Colin Fyfe

Abstract: We investigate the use of the new Cross Entropy method as a tool for exploratory data analysis. We show how this method can be used to perform linear projections such as principal component analysis, exploratory projection pursuit and canonical correlation analysis. We further go on to show how topology preserving mappings can be created usin the cross entropy method. We also show how the cross entropy method can be used to train deep architecture nets which are one of the main current research directions for creating true artificial intelligence. Finally we show how the cross entropy method can be used to optimize parameters for latent variable models.

Keywords: Cross entropy, Linear projections, Topographic mapping.

Title of the Paper:  Electrical Energy Consumption Forecasting Based on Cointegration and a Support Vector Machine in China


Authors: Zhang Xing-Ping, Gu Rui

Abstract: By undertaking a cointegration analysis with annual data over the period 1985~2005 in China, the estimation results show that there is cointegration relationship between electrical energy consumption and economic growth taking into account industry structure changes and technical efficiency. The model shows that three explanatory variables, the GDP per capita, heavy industry share and efficiency improvement are the crucial factors which influence the electric energy consumption. The three explanatory variables and the actual electrical energy consumption are input into a support vector machine(SVM), a Gaussian radial basis function is taken as the kernel function and electrical energy consumptions from 1994~2006 are forecasted. The forecast results prove that the multivariable SVM is valid in forecasting electrical energy consumption in China.

Keywords: Cointegration analysis; Electrical energy consumption; Johansen cointegration test; Multivariate time series; Support vector machine ; Unit root test

[Journals], [Books], [Conferences], [Research], [Contact us], [Indexing], [Reports], [E-Library], [FAQ],
[Upcoming Conferences with Expired Deadline], [History of the WSEAS Conferences]

Copyright © WSEAS