With the growing market demand for Dendrobium officinale

consumers are increasingly concerned about its quality. Distinct geographical environments lead to significant variations in the content of bioactive compounds such as polysaccharides and alkaloids

which in turn result in considerable differences in market value. However

the abundance of brands and uneven product quality

coupled with the prevalence of counterfeits

pose challenges to quality assurance. Chemometrics

with its strength in analyzing large datasets

has been widely applied in the geographical traceability of medicinal materials. In this study

near-infrared (NIR) spectroscopy combined with chemometric approaches including SVM

PLS-DA

and KNN was employed to develop origin discrimination strategies for D. officinale. By optimizing the training set proportion

the results revealed that the traceability model constructed using the DT preprocessing method in combination with KNN exhibited the best performance

achieving 100% classification accuracy. These findings provide a highly accurate and efficient technique for the geographical authentication of D. officinale.