MULTIVARIATE PATTERN RECOGNITION OF DIFFERENT VARIETIES' LAVENDER BY HEAD SPACE-SOLID PHASE MICROEXTRACTION WITH GAS CHROMATOGRAPHY

作者：Tang, Jun^*; Fu, Jihong; Tong, Hong; Liao, Xiang

来源：Bangladesh Journal of Botany, 2016, 45(3): 659-667.

• Lavender
• Head space-solid phase
• Microextraction
• Principal component analysis
• Partial least squares-discriminate analysis

Head space-solid phase microextraction (HS-SPME) coupled with gas chromatography (GC) has been applied for the identification of the characteristic volatile profile of lavender with the aim to study its varieties from Yili Xinjiang areas. Quantitative determinations of lavender samples' common peaks were carried out by GC with flame ionization detection (GC-FID) and qualitative analysis by GC with mass spectrometry (GC-MS). Principal components analysis (PCA) showed group clustering of three types of lavender varieties using GC-FID data. Partial least squares-discriminant analysis (PLS-DA) models had excellent classification sensitivity both for the calibration set and for the test set samples. The PLS-DA method was evaluated by the statistical indices of the correct recognition rate. Results showed that PLS-DA model were much better than PCA in classificatory abilities.. It could successfully identify the complex nonlinearity and correlations among input variables and minimize them. The proposed chemometric methods illustrating the very plummy multivariate classification models, it proved to be an effective strategy for identifying the varieties of lavender, especially give a feasible method in the lavender quality control for use in medicine.

• 出版日期2016-9
• 单位新疆大学