DocumentCode :
3457146
Title :
The Bayesian Network Method for Scenario Prediction
Author :
Shuang-cheng Wang ; Shao Jun ; Du Rui-jie
Author_Institution :
Sch. of Math. & Inf., Shanghai Lixin Univ. of Commerce, Shanghai, China
fYear :
2010
fDate :
21-23 Oct. 2010
Firstpage :
1
Lastpage :
5
Abstract :
Bayesian network is a powerful tool for scenario analysis and prediction. At present, future scenario is structured by combining the results of forecasting every variable. The dependency relationship between variables is neglected so that the result of scenario prediction often is unreliable. In this paper, a Bayesian network is built by combining subjective expert knowledge and objective data and a scenario is forecasted by joint probability calculation based on Bayesian network. Because the dependency relationship between variables can be effectively used, forecasting result will be more believable and prediction and analysis on scenario can also combined.
Keywords :
belief networks; expert systems; prediction theory; probability; Bayesian Network Method; expert knowledge; forecasting; joint probability calculation; scenario analysis; scenario prediction; Bayesian methods; Biological system modeling; Business; Data mining; Electronic mail; Learning; Load modeling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition (CCPR), 2010 Chinese Conference on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4244-7209-3
Electronic_ISBN :
978-1-4244-7210-9
Type :
conf
DOI :
10.1109/CCPR.2010.5659201
Filename :
5659201
Link To Document :
بازگشت