Machine learning for authentication of black tea from narrow-geographic origins - Combination of PCA and PLS with LDA and SVM classifiers

Aug 6, 2026

Upload by Eliana

This study investigates the feasibility of using UV–Vis spectroscopy coupled with machine learning methods to authenticate tea samples based on their geographical origins in a narrow longitudinal strip (200 km). Several preprocessing methods, including SNV, auto-scaling, MSC, MC, and first-derivative approaches, were applied to remove noninformative information. The PLS-LDA model using first-derivative spectra achieved 98.0% sensitivity, 99.5% specificity, and 98.0% mean accuracy, while the PLS-SVM classifier showed 94.0% sensitivity, 98.6% specificity, and 94.0% mean accuracy. Results indicate UV-absorbing chemical components can discriminate origin.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Tea quality and the role of geographical origin

Analytical methods for origin identification

Research motivation and objective

LWT-Food Science and Technology 203 (2024) 116401 Contents lists available at ScienceDirect LWT journal [homepage:](homepage: www.elsevier.com/locate/lwt)[ www.elsevier.com/locate/lwt](homepage: www.elsevier.com/locate/lwt) | Machine learning for authentication of black tea from narrow-geographic origins: Combination of PCA and PLS with LDA and SVM classifiers | | | | | --- | --- | --- | --- | | Nahid Mohammadia , Mahnaz Estekia, ** , Jesus Simal-Gandarab, *
a Department of Chemistry, University of Zanjan, Zanjan, 45195-313, Iran
b Universidade de Vigo, Nutrition and Bromatology Group, Analytical Chemistry and Food Science Department, Faculty of Science, E-32004, Ourense, Spain | | | | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords:
Machine learning Authentication
Tea
Geographical origin UV–Vis spectral fingerprint | | This study investigates the feasibility of using UV–Vis spectroscopy coupled with machine learning methods to authenticate tea samples based on their geographical origins in a narrow longitudinal strip (200 km). Several preprocessing methods, such as standard normal variate (SNV), auto-scaling, multiplicative scatter correction (MSC), mean centring (MC), first derivative, and their combinations, were applied to eliminate the noninformative information. The partial least squares-linear discriminant analysis (PLS-LDA) model using first derivative spectra represented the following results, including 98.0% sensitivity, 99.5% specificity, and a mean accuracy of 98.0%. The support vector machine (PLS-SVM) classifier using first derivative spectra represented 94.0% sensitivity, 98.6% specificity, and a mean accuracy of 94.0%. The satisfactory results of the models depicted that the chemical components of tea, such as polyphenols, chlorogenic and fatty acids that absorb UV radiation are the chemical markers that can discriminate tea samples based on their geographical origin. Therefore, UV–Vis spectral fingerprinting combined with machine learning methods could be a practical, feasible, and simple method for classifying tea based on their geographical origins in a narrow longitudinal strip. | | 1. Introduction Tea is the second most consumption beverage in the world, which its characteristic flavour is due to the chemical components such as catechins, vitamins, amino acids, caffeine, and volatile aroma components derived from tea leaves (Wen et al., 2021). This little evergreen tree was initially found by the Chinese and has been used mostly by Asian countries for many years. However, it is now produced in over 30 countries worldwide (Xia et al., 2020). The chemical composition of tea leaves determines their quality characteristics. This parameter is mainly influenced by geographical features and environmental factors such as slope aspect, altitude, soil characteristics, temperature, rainfall, and sun exposure (Wen et al., 2020). For example, the catechin content in tea leaves increases with increasing the time of sunshine (Ghabru et al., 2017). Furthermore, a higher concentration range of theanine and lower concentrations of valine, isoleucine, alanine, leucine, caffeine, and flavonoids are detected in tea leaves grown in areas with high rainfall, high temperature, and prolonged sun exposure time (Lee et al., 2010). Based on the above information, it can be concluded that the geographical origin of the tea samples may substantially impact their quality. Nowadays, several countries or geographical areas have the unique ability to produce high-quality tea, allowing them to market their products at premium prices. The aim to increase earnings has led other companies to promote their low quality items as high grade from these specific geographical regions by utilizing fake labelling (Jurica et al., 2021). This issue draws customers’ attention to the geographical origin of the tea they drink as a key indicator of its quality. Several techniques are available for the id

FAQs

What data source and spectral range are used for tea authentication?

UV–Vis spectroscopy is used, with absorbance spectra recorded from 220 to 420 nm for black tea samples.

Which preprocessing methods improve the UV–Vis spectral information?

Standard normal variate (SNV), auto-scaling, MSC, mean centring, first-derivative spectra, and their combinations were applied to reduce noninformative variation.

How did the PLS-LDA and PLS-SVM models perform in discriminating tea origins?

PLS-LDA with first-derivative spectra reached 98.0% sensitivity, 99.5% specificity, and 98.0% mean accuracy; PLS-SVM achieved 94.0% sensitivity, 98.6% specificity, and 94.0% mean accuracy.

Aug 6, 2026

Upload by Eliana