Rapid identification of the geographical origin of Eucommia ulmoides by using excitation-emission matrix fluorescence combined with chemometric methods

Eucommia ulmoides is an important and valuable traditional Chinese medicine with various medical functions, and has been widely used as health food in China, Japan, South Korea and other Asian countries for many years. The efficacy and quality of E. ulmoides are closely associated with the geographical origin. In this work, the potential of excitation-emission matrix (EEMs) fluorescence coupled with chemometric methods was investigated for simple, rapid and accurate for identification E. ulmoides from different geographical origins. Parallel factor analysis (PARAFAC) was applied for characterizing the fluorescence fingerprints of E. ulmoides samples. Moreover, k-nearest neighbor (kNN), principal component analysis-linear discriminant analysis (PCA-LDA) and partial least squares discriminant analysis (PLS-DA) models were used for the classification of E. ulmoides samples according to their geographical origins. The results showed that kNN model was more suitable for identification of E. ulmoides samples from different provinces. The kNN model could identify E. ulmoides samples from eight different geographical origins with 100% accuracy on the training and test sets. Therefore, the proposed method was available for conveniently and accurately determining the geographical origin of E. ulmoides, which can expect to be an attractive alternative method for identifying the geographic origin of other traditional Chinese medicines.