|
|||||
|
|
| 面向属性效应控制的等均值约束支持向量回归机 | |
| 引用本文： | 刘解放,王士同,王骏,邓赵红.面向属性效应控制的等均值约束支持向量回归机[J].控制与决策,2016,31(10):1791-1797. |
| 作者姓名： | 刘解放 王士同 王骏 邓赵红 |
| 作者单位： | 1. 江南大学数字媒体学院，江苏无锡214122； 2. 湖北交通职业技术学院交通信息学院，武汉430079. |
| 基金项目： | 国家自然科学基金项目(61272210, 61572236)；江苏省杰出青年基金项目(BK20140001)；江苏省自然科学基金项目(BK20130155, BK20151299). |
| 摘 要： | 针对现有的属性效应控制方法无法有效控制非线性回归建模的属性效应问题, 基于间隔最大化和结构风险最小化原则, 通过向SVR 目标学习准则中施加等均值条件约束, 提出等均值支持向量回归机(EM-SVR). 所提出的方法具有较好的泛化能力, 同时继承了EM-LS 的良好性能. 实验结果验证了所提出方法的有效性. |
| 关 键 词： | 支持向量回归机|属性效应控制|等均值约束|结构风险最小化 |
| 收稿时间： | 2015/10/24 0:00:00 |
| 修稿时间： | 2016/2/29 0:00:00 |
Equal mean based support vector regression for attribute effect control |
|
| LIU Jie-fang WANG Shi-tong WANG Jun DENG Zhao-hong.Equal mean based support vector regression for attribute effect control[J].Control and Decision,2016,31(10):1791-1797. | |
| Authors: | LIU Jie-fang WANG Shi-tong WANG Jun DENG Zhao-hong |
| Abstract: | In view of the problem that the existing attribute effect control method can not effectively control the attribute effect in the nonlinear regression model, an equal mean-support vector regression(EM-SVR) based on the principles of the margin maximization and the structural risk minimization is proposed by using the constraint condition of equal mean, which has the good generalization ability and the characteristic of nonlinear regression. At the same time, the good performance of equal mean-least square(EM-LS) is also inherited. Finally, the experiment results show the effectiveness of the proposed method. |
| Keywords: | support vector regression|attribute effect control|equal mean|structural risk minimization |
| 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 | |