| 加权最小二乘支持向量机与最小二乘问题的关系 |
| |
| 引用本文： | 陈将宏 周德强 李落清. 加权最小二乘支持向量机与最小二乘问题的关系[J]. 湖北大学学报(自然科学版), 2004, 26(1): 16-18,26 |
| |
| 作者姓名： | 陈将宏 周德强 李落清 |
| |
| 作者单位： | 湖北大学,数学与计算机科学学院,湖北,武汉,430062;湖北大学,数学与计算机科学学院,湖北,武汉,430062;湖北大学,数学与计算机科学学院,湖北,武汉,430062 |
| |
| 基金项目： | 湖北省自然科学基金(99J169)资助课题 |
| |
| 摘 要： | 研究了加权最小二乘支持向量机与最小二乘法的关系．证明了用加权最小二乘支持向量机作函数估计与在特征空间中用最小二乘法得到的解是一致的．加权最小二乘支持向量机选择核相当于最小二乘法选择基函数组．由此提出了采用加权最小二乘支持向量机解决最小二乘法问题的思想，保证解具有良好的推广性、鲁棒性与稀疏性．
|
| 关 键 词： | 支持向量机 最小二乘法 特征空间 |
| 文章编号： | 1000-2375(2004)01-0016-03 |
The relation between weighted least squares support vector machine and the least square problem |
| |
| CHEN Giang-hong.ZHOU De-qiang,LI Luo-qing. The relation between weighted least squares support vector machine and the least square problem[J]. Journal of Hubei University(Natural Science Edition), 2004, 26(1): 16-18,26 |
| |
| Authors: | CHEN Giang-hong.ZHOU De-qiang LI Luo-qing |
| |
| Abstract: | The relationship between weighted least square SVM and the least square method is studied. It is proved that the solution for regression by using weighted least squares SVM coincide with the solution obtained by using least square method in feature space. To choose the kernel in weighted least square SVM is equivalent to choosing the basis functions. A scheme for solving LS problems with weighted least squares SVM, which guarantees the good generalization performance of the solution and reinforces the robustness and sparseness of it, is presented. |
| |
| Keywords: | support vector machine least square method feature space |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
