| 基于SVM决策树判别测试点类别的新方法 |
| |
| 引用本文： | 薛欣,贺国平.基于SVM决策树判别测试点类别的新方法[J].计算机应用,2007,27(1):84-85. |
| |
| 作者姓名： | 薛欣 贺国平 |
| |
| 作者单位： | 山东科技大学,信息科学与工程学院,山东,青岛,266510;泰山学院,数学与系统科学系,山东,泰安271021;山东科技大学,信息科学与工程学院,山东,青岛,266510 |
| |
| 摘 要： | 从测试点的类别判断方式上进行改进，对容易错分的测试点给予多次判别机会，从而降低了SVM决策树的错分累积程度。仿真试验表明，改进的基于SVM决策树判别测试点类别方法与传统的基于SVM决策树判别测试点类别方法相比，具有较高的分类精度。
|
| 关 键 词： | 间隔 决策树 支持向量机 |
| 文章编号： | 1001-9081（2007）01-0084-02 |
| 收稿时间： | 2006-07-13 |
| 修稿时间： | 2006-07-13 |
New method for deciding the sort of the test sample based on SVM decision tree |
| |
| XUE Xin,HE Guo-ping.New method for deciding the sort of the test sample based on SVM decision tree[J].journal of Computer Applications,2007,27(1):84-85. |
| |
| Authors: | XUE Xin HE Guo-ping |
| |
| Abstract: | By giving the test sample many decision chances,a new method for deciding the sort of the test sample based on Support Vector Machine(SVM) decision tree was given.The new decision method reduces the degree of error accumulation.Experimental results show that the new decision method based on SVM decision tree yields higher precision than the traditional method. |
| |
| Keywords: | margin decision tree SVM(Support Vector Machine) |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机应用》浏览原始摘要信息 |
|
点击此处可从《计算机应用》下载全文 |
|
