| 混合核函数对支持向量机分类性能的改进 |
| |
| 引用本文： | 朱树先,张仁杰,郑刚. 混合核函数对支持向量机分类性能的改进[J]. 上海理工大学学报(社会科学版), 2009, 31(2) |
| |
| 作者姓名： | 朱树先 张仁杰 郑刚 |
| |
| 作者单位： | 上海理工大学,光电信息与计算机工程学院,上海,200093 ? |
| |
| 摘 要： | 通过对核矩阵的计算和研究,从理论上对常用的核函数进行了评估.在此基础上,通过实验仿真证实了通过优选后的核函数所组成的混合核函数对支持向量机分类性能的改善,为核函数的选择提供了参考.
|
| 关 键 词： | 支持向量机 核函数 模型选择 核矩阵 混合核函数 |
Using mixture of kernels to improve the classification performance of support vector machines |
| |
| ZHU Shu-xian,ZHANG Ren-jie,ZHENG Gang. Using mixture of kernels to improve the classification performance of support vector machines[J]. Journal of University of Shanghai For Science and Technilogy(Social Science), 2009, 31(2) |
| |
| Authors: | ZHU Shu-xian ZHANG Ren-jie ZHENG Gang |
| |
| Affiliation: | School of Optical-Electronial and Computer Engineering;University of Shanghai for Science and Technology;Shanghai 200093;China |
| |
| Abstract: | A reference for selecting the kernel function of support vector machine(SVM) theoretically was provided through observing and computing the kernel matrix.On this basis,the selected kernel functions were then used to get a mixture of kernels.Experiments and simalations prove that the performance of SVM does be improved by the mixed kernel function. |
| |
| Keywords: | support vector machine kernel function model selection kernel matrix mixture of kernels |
|
| 点击此处可从《上海理工大学学报(社会科学版)》浏览原始摘要信息 |
|
点击此处可从《上海理工大学学报(社会科学版)》下载全文 |
