| 基于RBF核的SVM分类应用研究 |
| |
| 引用本文： | 朱芳,赵庆平,王江涛.基于RBF核的SVM分类应用研究[J].唐山学院学报,2017,30(3):13-17,39. |
| |
| 作者姓名： | 朱芳 赵庆平 王江涛 |
| |
| 作者单位： | 淮北师范大学 物理与电子信息学院, 安徽 淮北 235000,淮北师范大学 物理与电子信息学院, 安徽 淮北 235000,淮北师范大学 物理与电子信息学院, 安徽 淮北 235000 |
| |
| 摘 要： | 支持向量机(Support Vector Machine,SVM)在解决小样本、非线性及高维模式识别中具有优势,但核函数的选取没有定论,且其参数对SVM模型的性能起重要作用。针对这些问题,文章建立了基于SVM的分类模型,并通过UCI数据集验证了径向基核函数(Radial Basis Function,RBF)较其他核函数的有效性,其中核参数的选取采用改进的网格搜索法进行寻优。分类实验结果表明,选择RBF核函数的分类准确度较其他核函数提高了2.5%到35%。
|
| 关 键 词： | 支持向量机 径向基核函数 网格法 分类 |
The Classification and Application of SVM Based on the RBF Kernel |
| |
| ZHU Fang,ZHAO Qing-ping and WANG Jiang-tao.The Classification and Application of SVM Based on the RBF Kernel[J].Journal of Tangshan College,2017,30(3):13-17,39. |
| |
| Authors: | ZHU Fang ZHAO Qing-ping and WANG Jiang-tao |
| |
| Institution: | School of Physics and Electrical Information, Huaibei Normal University, Huaibei 235000, China,School of Physics and Electrical Information, Huaibei Normal University, Huaibei 235000, China and School of Physics and Electrical Information, Huaibei Normal University, Huaibei 235000, China |
| |
| Abstract: | The support vector machine ( SVM) has some advantages in the small-sample, nonlinear and high dimensional pattern recognition, but the selection of kernel function is not conclusive, and its parameters have an important influence on the performance of the SVM model. To solve these problems, the authors of this paper established a classification model based on SVM, and verified the greater effectiveness of the radial basis function (Radial Basis Function, RBF) than that of other nuclear functions through the analysis of UCI data sets and kernel parameters were determined with the improved method of grid search. The experiment results show that the classification accuracy of RBF kernel function has been improved by 2.5% to 35% in comparison with other kernel functions. |
| |
| Keywords: | SVM RBF grid search method classification |
|
| 点击此处可从《唐山学院学报》浏览原始摘要信息 |
| 点击此处可从《唐山学院学报》下载免费的PDF全文 |
|
