| 一种快速稀疏最小二乘支持向量回归机 |
| |
| 引用本文： | 赵永平,孙健国.一种快速稀疏最小二乘支持向量回归机[J].控制与决策,2008,23(12):1347-1352. |
| |
| 作者姓名： | 赵永平 孙健国 |
| |
| 作者单位： | 南京航空航天大学,能源与动力学院,南京,210016 |
| |
| 基金项目： | 国家自然科学基金
|
| |
| 摘 要： | 将Jiao法直接应用于最小二乘支持向量回归机上的效果并不理想,为此采用不完全抛弃的策略,提出了改进的Jiao法,并将其应用于最小二乘支持向量回归机.数据集测试的结果表明,基于改进Jiao法的稀疏最小二乘支持向量回归机,无论在支持向量个数和训练时间上都取得了一定的优势.与其他剪枝算法相比,在不丧失回归精度的情况下,改进的Jiao法可大大缩短训练时间.另外,改进的Jiao法同样适用于分类问题.
|
| 关 键 词： | 最小二乘支持向量回归机 Jiao法 剪枝算法 分类 |
| 收稿时间： | 2007-9-22 |
| 修稿时间： | 2008-1-1 |
Fast method for sparse least squares support vector regression machine |
| |
| ZHAO Yong-ping,SUN Jian-guo.Fast method for sparse least squares support vector regression machine[J].Control and Decision,2008,23(12):1347-1352. |
| |
| Authors: | ZHAO Yong-ping SUN Jian-guo |
| |
| Affiliation: | ZHAO Yong-ping,SUN Jian-guo(College of Energy , Power Engineering,Nanjing University of Aeronautics , Astronautics,Nanjing 210016,China. |
| |
| Abstract: | Jiao's method is directly used for least squares support vector regression machine(LS-SVRM),but the effectiveness isn't satisfying.Therefore,an improved Jiao's method built on partial reduction is proposed,and a fast sparse least squares support vector regression machine(FPLS-SVRM) on the basis of the improved Jiao's method is presented.Simulation results of data sets demonstrate that FPLS-SVRM gains advantages over LS-SVRM based on Jiao's method in the number of support vectors and training time in a way.M... |
| |
| Keywords: | Least squares support vector regression machine Jiao's method Pruning algorithm Classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《控制与决策》浏览原始摘要信息 |
|
点击此处可从《控制与决策》下载全文 |
|
