作　　者： ; ; ; ; ;

机构地区： 首都师范大学化学系

出　　处： 《分析化学》 2008年第6期770-774,共5页

摘　　要： 采用一阶导数数据预处理,最小二乘支持向量机(LS-SVM)紫外可见光谱建模,对清开灵注射液四混中间体进行质量评价。以二次网格法和十折交叉验证法优化建模参数,预测集的总正确率和接受器工作特性曲线(ROC)下面积分别可达98.0%和0.983。结果表明,与经典的支持向量机相比,LSSVM鉴别准确率更高,模型的泛化能力更强。可用于清开灵注射液生产过程中质量控制,为中药注射液生产过程的质量控制提供了一条有效的途径。 The first derivative spectra with selected wavelengths were used to eliminate the slope-background and reduce variables for the measured ultraviolet （UV） spectra of Chinese medicinal Qingkailing injection intermediates. Then, least squares support vector machine （ES-SVM） was used for building the classification model to discriminate 196 injections intermediate samples （116 modeling parameters were investigated using two-grid searching qualified and 80 unqualified samples）. The and ten-fold cross-validation methods. Under the optimized conditions, the predictive ability of the testing set and the area under receiver operation charac- teristic （ROC） curves （AUR） reach 98.0% and 0.983, respectively. Comparing with the conventional support vector machine （SVM）,LS-SVM was found better accuracy and generalization. Results showed that LS-SVM technique can be a useful means for quality control of Chinese medicinal injection in the production process and other Chinese medicines.

关 键 词： 清开灵注射液 中间体 紫外光谱法 最小二乘支持向量机

领　　域： [理学] [自动化与计算机技术] [自动化与计算机技术]