基于线性判别分析的特征选择  

Feature selection based on linear discriminant analysis

作　　者：崔自峰[1] 吉小华 

机构地区：[1]中国电子科技集团公司第二十八研究所,南京210007 [2]南京市地方税务局计算机中心,南京210002

出　　处：《计算机应用》2009年第10期2781-2785,共5页journal of Computer Applications

摘　　要：提出一种新颖的基于特征抽取的特征选择方法,将特征选择问题建模为在子空间中的搜索问题,采用线形判别分析(LDA)的投影思想,对LDA施加一定的限制将其转换为对子空间的搜索优化问题,从而通过解LDA的优化问题得到特征选择的解,进一步把特征选择问题推导简化为对特征的评分和排序过程。通过在UC I机器学习库和Reuters-21578文本数据集上的实验,验证了该方法以较少的特征获得了比全部特征更好的分类结果。The paper proposed a new approach of feature selection based on Constrained Linear Discriminant Analysis （CLDA）, which modeled feature selection as a search problem in subspace and made optimal solution subject to some restrictions. Furthermore, CLDA optimization problem was transformed into a process of scoring and sorting features. Experiments on UCI machine learning repository and Reuters-21578 dataset show that the proposed approach can consistently obtain better results with fewer features than that with all features.

关 键 词：特征选择 线性判别分析 分类 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]