论文部分内容阅读
用支持向量机方法分析280个解表类中药挥发性成分的GC-MS数据,探讨解表药挥发性成分与药性的相关性。以水蒸汽蒸馏法提取各药材的挥发性成分,利用GC-MS方法对其挥发性成分进行分析:对各药材不同类型化学成分的含量进行分类统计,并以不同类型化学成分的含量统计结果作为药性分类的特征指标;采用交叉验证法,利用支持向量机对不同药性解表药的数据进行交叉训练,建立解表药药性的预测模型;该模型对预测集中的辛温类药的正确识别率为95.0%,对辛寒类药的正确识别率为91.7%,总正确率为93.6%。实验结果表明解表类中药挥发性成分与其寒热药性具有较高的相关性,其中以解表类中药中的脂肪族及脂肪酸类成分、单萜氧化物成分对识别辛凉解表与辛温解表两种药性的贡献率最大。
The support vector machine method was used to analyze the GC-MS data of 280 volatile compounds in Chinese herbal medicines for relieving the symptoms of Chinese herbal medicines. Volatile components of each medicinal material were extracted by steam distillation, and their volatile components were analyzed by GC-MS method. The contents of different types of chemical components in each medicinal material were classified and statistically analyzed. The statistical results of different types of chemical components Cross-validation method, the use of support vector machine for cross-training of different pharmacological solutions of drug data to establish a predictive model of drug resistance; this model for the prediction of the correct concentration of Xin Wen drug identification rate Was 95.0%. The correct identification rate of Xin-Han medicines was 91.7%, the total correct rate was 93.6%. The experimental results show that there is a high correlation between the volatile components of Chinese herbal medicines and their cold-heat medicinal properties. Among them, the aliphatic and fatty acid components and the monoterpene oxide components in the Chinese herbal medicines for the identification of Xinliangjie Table and Xinxin Solution Table two kinds of medicine the contribution of the largest.