| 基于支持向量机和遗传算法的特征选择 |
| |
| 引用本文： | 韦振中,黄廷磊. 基于支持向量机和遗传算法的特征选择[J]. 广西工学院学报, 2006, 17(2): 18-21,25 |
| |
| 作者姓名： | 韦振中 黄廷磊 |
| |
| 作者单位： | 1. 桂林电子科技大学,计算机系,广西,桂林,541004;广西工学院,信息与计算科学系,广西,柳州,545006
2. 桂林电子科技大学,计算机系,广西,桂林,541004 |
| |
| 摘 要： | 支持向量机是一个具有多类分类能力的非线性分类器,利用特征集中的不同特征组合构成特征子集,所提供的数据在支持向量机上的平均正确分类率为目标函数值,利用遗传算法在整个特征组合空间中搜索能实现平均正确分类率最大化的最优特征子集。
|
| 关 键 词： | 支持向量机 遗传算法 特征选择 |
| 文章编号： | 1004-6410（2006）02-0018-04 |
| 收稿时间： | 2006-03-29 |
| 修稿时间： | 2006-03-29 |
Feature selection based on SVM and GA |
| |
| WEI Zheng-zhong,HUANG Ting-lei. Feature selection based on SVM and GA[J]. Journal of Guangxi University of Technology, 2006, 17(2): 18-21,25 |
| |
| Authors: | WEI Zheng-zhong HUANG Ting-lei |
| |
| Affiliation: | 1. Department of Computer Science, Guilin University of Electronic Technology, Guilin 541004, China; 2. Department of Information and Computation Science, Guangxi University of Technology,Liuzhou 545006,China |
| |
| Abstract: | Support vector machine is a nonlinear classifier with multi-classification performances.In this paper,we get different features from the feature set and combine them as a feature subset.Then,we use the data of this subset to measure the mean correct classification rate by means of SVM classifier and regard the result as the objective function value of the feature subset.Finally,we search the optimal feature subset to realize the maximization of the mean correct classification rate by using the genetic algorithm. |
| |
| Keywords: | support vector machines genetic algorithms feature selection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
