| 基于改进粒子群算法的最优特征子集研究 |
| |
| 引用本文： | 侯大军,朱伟兴. 基于改进粒子群算法的最优特征子集研究[J]. 传感器与微系统, 2010, 29(9) |
| |
| 作者姓名： | 侯大军 朱伟兴 |
| |
| 作者单位： | 江苏大学电气信息工程学院; |
| |
| 摘 要： | 特征选择是模式识别系统的难点.针对高维数据对象,先运用改进粒子群优化(PSO)算法快速、有效地从特征样本中提取一组最优特征子集,然后采用最小二乘支持向量机(LSSVM)分类器对最优特征子集进行分类,验证特征选择的好坏.经大量实验验证,在保证分类正确率的前提下,该方法有效提高了特征选择效率.
|
| 关 键 词： | 特征选择 粒子群优化算法 最小乘支持向量机 |
Optimization of a subset of features based on modified particle swarm optimization |
| |
| HOU Da-jun,ZHU Wei-xing. Optimization of a subset of features based on modified particle swarm optimization[J]. Transducer and Microsystem Technology, 2010, 29(9) |
| |
| Authors: | HOU Da-jun ZHU Wei-xing |
| |
| Affiliation: | HOU Da-jun,ZHU Wei-xing (School of Electrical and Information Engineering,Jiangsu University,Zhenjiang 212003,China) |
| |
| Abstract: | Feature selection is a very difficult problems for mode recognition system.For high dimension data,the modified particle swarm optimization(PSO) algorithm was effectively and quickly applied to the feature extraction of the optimum samples.The optimum samples were recognized by the support vector machine classifier,which is checked how to use the feature selection.Experimental results on the database show this method can guarantee the correct rate of classification and improve the efficiency of feature sele... |
| |
| Keywords: | feature selection particle swarm optimization(PSO)algorithm least squares support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
