| 基于核的k最近邻算法 |
| |
| 引用本文： | 张瑞民,郭海刚,李培峦.基于核的k最近邻算法[J].华北水利水电学院学报,2007,28(6):51-52. |
| |
| 作者姓名： | 张瑞民 郭海刚 李培峦 |
| |
| 作者单位： | 河南科技大学理学院,河南,洛阳,471003 |
| |
| 基金项目： | 河南科技大学校科研和教改项目 |
| |
| 摘 要： | 将核学习的方法应用到k最近邻算法中,提出了1种基于核的k最近邻算法.该算法通过引入核函数,将原空间中的样本映射到1个高维核空间中,突出了不同类别样本之间的特征差异,使样本变得线性可分,从而提高分类性能.并利用UCI数据库中的数据分别验证了k最近邻分类和核k最近邻分类算法的有效性.
|
| 关 键 词： | k最近邻算法 核方法 核函数 分类 |
| 文章编号： | 1002-5634（2007）06-0051-02 |
| 收稿时间： | 2007-08-12 |
| 修稿时间： | 2007-10-09 |
kNN Algorithm Based on Kernel Methods |
| |
| ZHANG Rui-min,GUO Hai-gang,LI Pei-luan.kNN Algorithm Based on Kernel Methods[J].Journal of North China Institute of Water Conservancy and Hydroelectric Power,2007,28(6):51-52. |
| |
| Authors: | ZHANG Rui-min GUO Hai-gang LI Pei-luan |
| |
| Abstract: | The idea of kernel based on learning method is applied to kNN algorithms and an algorithm of kernel kNN algorithms is proposed. By using kernel function,the data in the original space is mapped into a high dimensional kernel space,where feature differences among all kinds of samples are stood out, so that classifying could be performed more correctly. The performance of new algorithm is demonstrated to be superior to that of kNN algorithm by experiments on real data from UCI. |
| |
| Keywords: | kNN algorithm kernel methods kernel function classification |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
