|
|||||
|
|
| 基于偏最小二乘特征提取的支持向量机回归算法 | |
| 引用本文： | 刘杰,甘旭升,高海龙,王美乂.基于偏最小二乘特征提取的支持向量机回归算法[J].火力与指挥控制,2009,34(9). |
| 作者姓名： | 刘杰 甘旭升 高海龙 王美乂 |
| 作者单位： | 1. 西京学院,陕西,西安,710123 2. 空军工程大学工程学院,陕西,西安,710038 3. 解放军93132部队,黑龙江,齐齐哈尔,161000 4. 西安军代局驻203所军代室,陕西,西安,710065 |
| 摘 要： | 为了提高SVM的建模质量,简化建模难度,提出了PLS-SVM组合回归建模方法.该方法通过PLS对样本数据进行降维、去噪以及消除共线性处理后,再进行SVM回归建模.不仅保持了SVM良好的模型性能,而且使SVM具备特征提取功能.实验结果表明,该方法是可行的,采用此法构建的SVM模型,泛化性能优于没有特征提取的SVM. |
| 关 键 词： | 特征提取 支持向量机 偏最小二乘 主成分 |
A Support Vector Machine Regression Algorithm based on Partial Least Squares Feature Extraction |
|
| LIU Jie,GAN Xu-sheng,GAO Hai-long,WANG Mei-yi.A Support Vector Machine Regression Algorithm based on Partial Least Squares Feature Extraction[J].Fire Control & Command Control,2009,34(9). | |
| Authors: | LIU Jie GAN Xu-sheng GAO Hai-long WANG Mei-yi |
| Institution: | 1.Xijing College;Xi'an 710123;China;2.The Engineering Institute;Air Force Engineering University;Xi'an 710038;3.93132th Unit of the PLA;Qiqihar 161000;4.The Section of PLA Representation in 203th Troop of the PLA;Xi'an 710065;China |
| Abstract: | A hybrid PLS-SVM method is proposed to improve the SVM model quality and reduce the modeling difficulty.Firstly,reduced the dimensions of correlated inputs and denoised for sample by PLS,then construct the SVM model.The PLSSVM not only maintains the SVM good performance but also has feature extraction function.The experiment results show that this method is workable we11 and the generalization ability of SVM with feature extraction using PLS is much better than that without feature extraction. |
| Keywords: | feature extraction support vector machines partial least square principal component |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |