DocumentCode
2258392
Title
Prediction Discrete Data Used BP Network Based on AGO
Author
Gao, Xuejun ; Wang, Zhenyou
Author_Institution
Coll. of Math., Guangdong Univ. of Technol., Guangzhou, China
fYear
2010
fDate
11-14 Dec. 2010
Firstpage
218
Lastpage
222
Abstract
A method for prediction discrete data is presented in this article. In order to forecast the discrete data, the experiment that use the GM (1,1) and BP networks to predict discrete data are respectively executed, we found that AGO operation in the GM method can effectively reduce randomness of the discrete data, so AGO operation is applied into the BP network method. According to the result of the simulation experiment, we have obtained more significantly and effectively results than BP network with no used AGO operation. It is explained that BP network can effectively forecast the discrete data with AGO operation.
Keywords
backpropagation; data analysis; forecasting theory; neural nets; AGO; BP network; GM method; accumulation generation operation; discrete data prediction; gray model; BP network; accumulation generation operation (AGO); prediction discrete data;
fLanguage
English
Publisher
ieee
Conference_Titel
Computational Intelligence and Security (CIS), 2010 International Conference on
Conference_Location
Nanning
Print_ISBN
978-1-4244-9114-8
Electronic_ISBN
978-0-7695-4297-3
Type
conf
DOI
10.1109/CIS.2010.53
Filename
5696266
Link To Document