首页 | 本学科首页   官方微博 | 高级检索  
     

基于SVM的特征筛选方法及其若干应用
引用本文:李国正,王振晓,杨杰,姚莉秀,陈念贻.基于SVM的特征筛选方法及其若干应用[J].计算机与应用化学,2002,19(6):703-705.
作者姓名:李国正  王振晓  杨杰  姚莉秀  陈念贻
作者单位:1. 上海交通大学图象及模式识别研究所,上海,200030
2. 上海大学理学院化学系计算机化学研究室,上海,200436
基金项目:由国家自然科学基金和上海宝钢集团公司联合资助(50174038)
摘    要:对于拟合问题,传统的模式识别特征筛选方法以各特征量对训练数据拟合能力的贡献为取舍标准,未考虑经验风险最小化和结构风险最小化间的差别,不能获得预报能力最强的特征筛选结果。为此我们提出了结合支持向量回归法与留一法的特征筛选新算法,并将它试用于镍氢电池材料和氧化铝溶出率两套实验数据集的特征筛选。

关 键 词:SVM  应用  特征筛选  支持向量回归  留一法  预报能力  化学模式识别  镍-氢电池  材料  电化学容量  净溶出率  氧化铝
文章编号:1001-4160(2002)06-703-705
修稿时间:2002年9月16日

A SVM-based feature selection method and its applications
LI GUO-zheng,WANG Zheng-xiao,YANG Jie,YAO Li-xiu,CHEN Nian-yi.A SVM-based feature selection method and its applications[J].Computers and Applied Chemistry,2002,19(6):703-705.
Authors:LI GUO-zheng  WANG Zheng-xiao  YANG Jie  YAO Li-xiu  CHEN Nian-yi
Abstract:Most of the traditional feature selection methods designed for regression problem only consider the consistence between the training data and the regression result. However, they neglect the vital differences betvveen the empirical risk minimization and the structural risk minimization and hence cannot directly find the feature subset with high generalization performance. To solve this problem, this paper proposed a floating search method for feature selection, which was based on support vector regression and leaving-one method. Experiments were carried out on two chemical data sets.
Keywords:feature selection  support vector regression  leaving-one method  generalization performance
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号