Synchronously Predicting Tea Polyphenol and Epigallocatechin Gallate in Tea Leaves Using Fourier Transform-Near-Infrared Spectroscopy and Machine Learning

This study aimed to develop reliable models for predicting tea polyphenols and epigallocatechin gallate (EGCG) content in tea leaves using Fourier Transform-near-infrared (FT-NIR) spectroscopy and machine learning. Various spectral preprocessing methods were applied, and partial least squares regression (PLSR) and least squares support vector regression (LS-SVR) were used to build predictive models. Variable selection algorithms (CARS and RF) further improved model efficiency. The optimal model for tea polyphenols was LS-SVR with R_p = 0.975 and RPD = 4.540, while for EGCG, it was LS-SVR with R_p = 0.936 and RPD = 2.841. Variable selection enhanced prediction, reaching R_p = 0.978 and RPD = 4.833 for tea polyphenols and R_p = 0.944 and RPD = 3.049 for EGCG. The results demonstrate the potential of FT-NIR spectroscopy combined with machine learning for rapid tea polyphenol and EGCG content screening.

Sitan Ye, Haiyong Weng, Lirong Xiang, Liangquan Jia, Jinchai Xu