Title :
A Comparative Study of Three Smooth SVM Classifiers
Author :
Xiong, Jinzhi ; Hu, Tianming ; Li, Guangming ; Peng, Hong
Author_Institution :
Software Coll., Dongguan Univ. of Technol.
Abstract :
Researching smooth support vector machine (SVM) for classification is an active field in data mining. This paper presents a comparison among three smooth SVM classifiers, smooth SVM, 1st-order polynomial smooth SVM and 2nd-order polynomial smooth SVM. Numerical experiments are performed to compare accuracy and computational complexity of these classifiers with linear and nonlinear kernels. This study provides some guidelines for future research and choosing an appropriate smooth SVM for classification
Keywords :
computational complexity; pattern classification; polynomials; support vector machines; 1st-order polynomial smooth SVM; 2nd-order polynomial smooth SVM; computational complexity; data mining; linear kernels; nonlinear kernels; smooth support vector machine classifiers; Computational complexity; Computer science; Data engineering; Data mining; Educational institutions; Kernel; Polynomials; Smoothing methods; Support vector machine classification; Support vector machines; classification; data mining; smoothing; support vector machine;
Conference_Titel :
Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on
Conference_Location :
Dalian
Print_ISBN :
1-4244-0332-4
DOI :
10.1109/WCICA.2006.1714223
