The server is under maintenance between 08:00 to 12:00 (GMT+08:00), and please visit later.
We apologize for any inconvenience caused
Login  | Sign Up  |  Oriprobe Inc. Feed
China/Asia On Demand
Journal Articles
Bookmark and Share
Thin wall tube method based sparse least squares support vector regression
Pages: 581-585
Year: Issue:  4
Journal: Electric Machines and Control

Keyword:  support vector machinesleast squares approximationssparsenesspruning algorithmsthin wall tube method;
Abstract: The solution of least squares support vector regression (LS-SVR) is lack of sparseness. To this end, after combing with the method of selecting support vectors in support vector classifier, the thin wall tube method (TWTM) is proposed. TWTM constructed a hollow tube with the finite wall through learning errors, which was different from the traditional pruning algorithms which select support vectors through infinite wall tubes so that outliers are not oppressed. Compared with the existing pruning algorithms, TWTM did not only reduce the number of support vectors (NSV) and curtail predictive time obviously, but also oppressed outliers existing in the system and enhanced the predictive accuracy to a extent. Experiments on simulation example show the effectiveness of the proposed TWTM.
Related Articles
No related articles found