首页 | 本学科首页   官方微博 | 高级检索  
     检索      

结构描述符正交化及典型相关分析在饱和醇、醚质谱分类中的应用
引用本文:贠栋,梁逸曾,徐承建,胡黔楠,梁江龙.结构描述符正交化及典型相关分析在饱和醇、醚质谱分类中的应用[J].应用化学,2004,21(3):223-0.
作者姓名:贠栋  梁逸曾  徐承建  胡黔楠  梁江龙
作者单位:中南大学化学化工学院化学计量学及智能仪器研究所,长沙,410083
基金项目:国家自然科学基金重点项目 (2 0 2 3 5 0 2 0,2 0 175 0 3 6)
摘    要:结构描述符正交化及典型相关分析在饱和醇、醚质谱分类中的应用;饱和醇醚;模式识别;质谱分类变量;块变量;典型相关分析

关 键 词:结构描述符正交化及典型相关分析在饱和醇、醚质谱分类中的应用  饱和醇醚  模式识别  质谱分类变量  块变量  典型相关分析
文章编号:1000-0518(2004)03-0223-04
收稿时间:2009-06-29
修稿时间:2003年6月13日

Application of Orthogonal Computation of Block Variables and Canonical Correlation Analysis in Mass Spectral Classification of Saturated Alcohol and Ether
YUN Dong,LIANG Yi-Zeng,XU Chen-Jian,HU Qian-Nan,LIANG Jiang-Long.Application of Orthogonal Computation of Block Variables and Canonical Correlation Analysis in Mass Spectral Classification of Saturated Alcohol and Ether[J].Chinese Journal of Applied Chemistry,2004,21(3):223-0.
Authors:YUN Dong  LIANG Yi-Zeng  XU Chen-Jian  HU Qian-Nan  LIANG Jiang-Long
Institution:YUN Dong,LIANG Yi-Zeng*,XU Chen-Jian,HU Qian-Nan,LIANG Jiang-Long
Abstract:The chemical pattern recognition method has been applied to 377 kinds of saturated ethers and alcohols in NIST62 mass spectral database. Based on block variables thought, available information about classification was drawn from several series of classifers by orthogonal computation and canonical correlation analysis(CCA), the discriminant least square models were thus set up to classify these ether and alcohol samples. The error probability for optimum model was about 1.06%, with cross validation error probability being about 1.36%.
Keywords:mass spectral classification  saturated alcohol and ether  pattern recognition  mass classifier  block variables  canonical correlation analysis
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《应用化学》浏览原始摘要信息
点击此处可从《应用化学》下载免费的PDF全文
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