Title :
Study of a new online Least Squares Support Vector Machine algorithm in gas prediction
Author :
Zhao, Xiao-hu ; Zhao, Ke-ke
Author_Institution :
Sch. of Commun. & Electron. Eng., China Univ. of Min. & Technol., Beijing
Abstract :
This paper studied on time series prediction, and proposes a new prediction algorithm of LS-SVM online learning against the shortcomings in the traditional online learning with least squares support vector machine. This algorithm was researched and used in coal mine gas prediction and had proved effective, compared with the actual data and other relative algorithms.
Keywords :
coal; least squares approximations; mining industry; prediction theory; support vector machines; time series; coal mine gas prediction; online least squares support vector machine algorithm; time series prediction algorithm; Geology; Least squares methods; Machine learning; Prediction algorithms; Prediction methods; Predictive models; Production; Safety; Support vector machines; Time series analysis; LS-SVM; gas; online prediction;
Conference_Titel :
Industrial Informatics, 2008. INDIN 2008. 6th IEEE International Conference on
Conference_Location :
Daejeon
Print_ISBN :
978-1-4244-2170-1
Electronic_ISBN :
1935-4576
DOI :
10.1109/INDIN.2008.4618253
