基于碳、氮稳定同位素特征的国产绿茶产地溯源研究

Study on the origin tracing of Chinese green tea based on the carbon and nitrogen stable isotope characteristics

中文关键词:  稳定同位素比率  绿茶  产区  溯源

英文关键词:carbon and nitrogen stable isotope ratios  green tea  producing regions  traceability

中文摘要:

      目的 研究我国多省出产绿茶碳、氮稳定同位素比率特征。优化基于稳定同位素比率差异的绿茶产地溯源模型。方法 运用同位素质谱检测我国多省出产的169份代表性绿茶样本。采用方差分析比较各产地样本是否具有显著性差异。按7:3划分训练集和测试集, 使用训练集构建基于线性判别分析(linear discriminant analysis, LDA)、支持向量机(support vector machine, SVM)和K-最近邻算法(K-nearest neighbor algorithm, KNN)的产地溯源判别模型。采用K折交叉验证的方法进行模型内部验证, 采用测试集进行模型外部验证。结果 各地绿茶样品的碳、氮稳定同位素比率存在显著性差异。建立基于LDA、SVM、KNN的判别模型, 对测试集预测的整体准确率分别为76.5%、86.3%和84.3%。非线性模型对此样本的总体判别效果优于线性模型。结论 碳、氮稳定同位素比率特征结合非线性判别模型可以用于绿茶产地溯源。本研究为绿茶溯源及判别模型参数优化提供了一定的参考。

英文摘要:

      Objective To study the carbon and nitrogen stable isotope ratio characteristics of green tea produced in multiple provinces in China, optimize the origin traceability model of green tea based on stable isotope ratio differences. Methods The 169 representative green tea samples from many provinces in China were detected by isotope mass spectrometry. Variance analysis was used to compare whether there were significant differences in samples from different producing areas. Divide the training set and test set by 7:3. The training set was used to construct the origin traceability discriminant model based on linear discriminant analysis (LDA), support vector machine (SVM) and K-nearest neighbor algorithm (KNN). The K-fold cross validation method was used for internal validation of the model, and the test set was used for external validation of the model. Results There were significant differences in the stable isotope ratios of carbon and nitrogen in green tea samples from different regions. The discriminant models based on LDA, SVM and KNN were established, and the overall accuracy of the test set prediction was 76.5%, 86.3% and 84.3%. The overall discriminant effect of nonlinear model on this sample was better than that of linear model. Conclusion Carbon and nitrogen stable isotope ratio characteristics combined with nonlinear discriminant model can be used to trace the origin of green tea. This study provides a reference for the traceability of green tea and the optimization of discriminant model parameters.