| 支持向量机分类算法研究 |
| |
| 引用本文： | 周宽久,张世荣.支持向量机分类算法研究[J].计算机工程与应用,2009,45(1):159-162. |
| |
| 作者姓名： | 周宽久 张世荣 |
| |
| 作者单位： | 大连理工大学,软件学院,辽宁,大连,116620 |
| |
| 摘 要： | 支持向量机在处理两类分类问题时,当两类样本混杂严重时会降低分类精度。在NN-SVM分类算法的基础上,通过计算样本点与其最近邻点类别的异同以及该点与其k个同类近邻点在核空间的平均距离修剪混淆点,进而提出了一种改进的NN-SVM算法——KCNN-SVM。实验数据表明,KCNN-SVM算法与SVM以及NN-SVM相比,有着更高的分类精度和更快的训练、分类时间。
|
| 关 键 词： | 支持向量机 核空间 文本分类 |
| 收稿时间： | 2007-12-26 |
| 修稿时间： | 2008-3-18
|
Support vector machines based classification algorithm |
| |
| ZHOU Kuan-jiu,ZHANG Shi-rong.Support vector machines based classification algorithm[J].Computer Engineering and Applications,2009,45(1):159-162. |
| |
| Authors: | ZHOU Kuan-jiu ZHANG Shi-rong |
| |
| Affiliation: | ZHOU Kuan-jiu,ZHANG Shi-rong Software School,Dalian University of Technology,Dalian,Liaoning 116620,China |
| |
| Abstract: | The accuracy of classification of SVM in a two-class classification problem would be decreased because of those promiscuous samples.KCNN-SVM is proposed in this paper as an improved NN-SVM algorithm，which prunes a sample according to their nearest neighbor’s class label as well as the average distance in kernel space between it and its k congener nearest neighbors.Experimental results show that KCNN-SVM algorithm is better than both SVM and NN-SVM in accuracy of classification and the total training and testing time is comparative to that of NN-SVM. |
| |
| Keywords: | Support Vector Machine kernel space text categorization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 |
|
点击此处可从《计算机工程与应用》下载全文 |
|
