Abstract:A fast,convenient,and efficient method was established for tracing the origin of bulbus of Fritillariae,thus better distinguishing the quality of bulbus of Fritillariae and promoting the development of its medicinal and food industry.Fritillaria cirrhosa D.Don,Fritillaria ussuriensis Maxim,Fritillaria pallidiflora Schrenk and Fritillaria thunbergii Miq were taken as research objects.A benchtop near-infrared spectrometer and chemometrics were employed to conduct spectral preprocessing.Four machine learning models—partial least squares-discriminant analysis（PLS-DA）,K-nearest neighbors （KNN）,support vector machine （SVM）,and decision tree （DT）—were adopted to construct an efficient and accurate classification model for tracing the origin.The results showed that the PLS-DA model performed the best,with the average accuracy of 100.00%,followed by SVM,and both KNN and DT had the accuracy over 85.00%.