补益中药材药效计算机分类法研究  

Study on Classfication of Efficacy of Benitit Traditional Chinese Medicine by Computer Pattern Recognition

作　　者：郑育平[1] 

机构地区：[1]广东医学院附属医院,广东湛江524001

出　　处：《广东微量元素科学》2004年第3期31-36,共6页Trace Elements Science

摘　　要：根据计算机模式识别法的马氏 (Mahalanobis)距离判别法原理 ,用VisualBasic语言编制应用程序软件系统 ,对补气、补阳、补阴和补血等四类补益中药材中的十二种矿物元素谱进行分类判别研究。当选择Sr、Cr、Co、Ni、Mn、Cu、Zn、Fe、Na、Ca等元素为特征参量 ,将药材中这些无机元素含量值作为客观指标 ,可准确地对四类补益药进行分类鉴别。训练样本的学习准确识别率为 94 7%,待判补益中药的准确预报率为 90 4%。本法为使用客观的检测指标 ,对中药材药效进行分类、鉴别提供了新的途径。Based on 12 mineral elements table,the classfication of the efficacy of benitit traditional Chinese medicine(TCM)was studied by computer pattern recognition(Mahalamobis distance desicion metnod).The result showed that when Sr,Cr,Co,Ni,Mn,Cu,Zn,Fe,Na,Ca were selected as the character parameter,the correct rate of judgement in training sample and in differentiated benifit TCM were 94 7% and 90 4% respectively.

关 键 词：补益中药材 矿物元素 计算机模式识别技术 药效分类判别 

分 类 号：R282.5[医药卫生—中药学] O235[医药卫生—中医学]