Discrimination of the Production Season of Chinese Green Tea by Chemical Analysis in Combination with Supervised Pattern Recognition

• Published in: Journal of agricultural and food chemistry Vol. 60; no. 28; pp. 7064 - 7070
• Main Authors: Xu, Wenping, Song, Qiushuang, Li, Daxiang, Wan, Xiaochun
• Format: Journal Article
• Language: English
• Published: Washington, DC American Chemical Society 18.07.2012
• Subjects: Amino Acids - analysis; Bayes Theorem; Bayesian theory; Biological and medical sciences; caffeine; Caffeine - analysis; Camellia sinensis - chemistry; Catechin - analysis; chemical analysis; China; Chromatography, High Pressure Liquid; Coffee, tea and other stimulative beverage industries; Discriminant Analysis; flavanols; Food industries; free amino acids; Fundamental and applied biological sciences. Psychology; green tea; high performance liquid chromatography; leaves; Seasons; spring; Tea - chemistry; China; green tea; K-nearest neighbor method; chemical composition; tea production season; pattern recognition; Chemical analysis; Discrimination; Production; Season; Green tea
• ISSN: 0021-8561; 1520-5118
• DOI: 10.1021/jf301340z

High-performance liquid chromatography (HPLC) has been used to quantify levels of free amino acids, catechins, and caffeine in Chinese green tea. Levels of free amino acids and catechins in green tea leaves show obvious variation from spring to summer, which is useful information to identify the production season of commercial green tea. Supervised pattern recognition methods such as the K-nearest neighbor (KNN) method and Bayesian discriminant method (a type of linear discriminant analysis (LDA)) were used to discriminate between the production seasons of Chinese green tea. The optimal accuracy of the KNN method was ≤97.61 and ≤94.80% as validated by resubstitution and cross-validation tests, respectively, and that of LDA was ≤95.22 and ≤93.54%, respectively. Compared with LDA, the KNN method did not require a Gaussian distribution and was more accurate than LDA. The KNN method in combination with chemical analysis is recommended for discrimination of the production seasons of Chinese green tea.