Authentication of Cassia seeds on the basis of two-wavelength HPLC fingerprinting with the use of chemometrics

High performance liquid chromatographic (HPLC) fingerprints of Cassia seed, a traditional Chinese medicine (TCM), were developed by means of the chromatograms at two wavelengths of 238 and 282 nm. Then, the two data sets were combined into one matrix. The application of principal component analysis (PCA) for this data matrix showed that the samples were clustered into four groups in accordance with the plant sources and preparation procedures. Furthermore, partial least squares (PLS), back propagation artificial neural network (BP-ANN), and radial basis function artificial neural network (RBF-ANN) were effectively applied to predict the category of the four different samples in the test set.