期刊名称：Journal of Computer Sciences and Applications
印刷版ISSN：2328-7268
电子版ISSN：2328-725X
出版年度：2016
卷号：3
期号：6
页码：181-184
DOI：10.12691/jcsa-3-6-14
出版社：Science and Education Publishing
摘要：One of the major problems in the study of Support vector machine (SVM) is kernel selection, that’s based necessarily on the problem of deciding a kernel function for a particular task and dataset. By contradiction to other machine learning algorithms, SVM focuses on maximizing the generalisation ability, which depends on the empirical risk and the complexity of the machine. We were focused on SVM trained using linear, polynomial, puk and Radial Basic Function (RBF) kernels. A preliminary study has been made between SVM using the best choice of kernel. Results had revealed that SVM trained using Linear Kernel is the best choice for dealing with Diabetes dataset.
关键词：data mining; machine learning algorithms; support vector machine; kernels function
