Tracing Geographical Origins of Teas Based on FT-NIR Spectroscopy: Introduction of Model Updating and Imbalanced Data Handling Approaches.

Xue-Zhen Hong,Zi-Hong Ye,Xian-Shu Fu,Zheng-Liang Wang,Li Zhang,Xiao-Ping Yu

doi:10.1155/2019/1537568

Abstract

This work presents a reliable approach to trace teas' geographical origins despite changes in teas caused by different harvest years. A total of 1447 tea samples collected from various areas in 2014 (660 samples) and 2015 (787 samples) were detected by FT-NIR. Seven classifiers trained on the 2014 dataset all succeeded to trace origins of samples collected in 2014; however, they all failed to predict origins for the 2015 samples due to different data distributions and imbalanced dataset. Three outlier detection based undersampling approaches—one-class SVM (OC-SVM), isolation forest and elliptic envelope—were then proposed; as a result, the highest macro average recall (MAR) for the 2015 dataset was improved from 56.86% to 73.95% (by SVM). A model updating approach was also applied, and the prediction MAR was significantly improved with increase in the updating rate. The best MAR (90.31%) was first achieved by the OC-SVM combined SVM classifier at a 50% rate.

Highlights

Tea is one of the most widely consumed beverages in the world because of its pleasurable taste, aroma, and healthy effects [1]
Driven by profits, special local teas are often replaced by inferior teas from other origins in the tea market. erefore, it is of significant importance to develop reliable geographical origin tracing approaches for teas
NIR responses to the four tea groups collected from different geographical origins and harvest years are presented in Figure 1, where the x-axis represents the spectra scanning range, and the y-axis represents the NIR response. e wavenumbers from 9000 to 12000 cm−1 were excluded due to their lower sensitivity and signal-to-noise ratio [14]

Summary

Introduction

Tea is one of the most widely consumed beverages in the world because of its pleasurable taste, aroma, and healthy effects [1]. Harvested tea leaves are processed differently to produce specific types of tea, such as unfermented green tea, fully fermented black tea and semifermented oolong tea. Oolong has been proven to be able to reduce obesity and control diabetes [2]. Ese factors would affect chemical compositions [3] of teas and determine their special aroma and taste—eventually their market prices. Many countries have published relevant regulations such as geographical indication (GI) certification to protect valuable products originated from certain areas. As a GI product, the famous Wuyi rock-essence (WY) tea only grows in certain areas of Wuyishan city, Fuijan province of China (GB/T 18745–2006) [4]. Erefore, it is of significant importance to develop reliable geographical origin tracing approaches for teas Driven by profits, special local teas are often replaced by inferior teas from other origins in the tea market. erefore, it is of significant importance to develop reliable geographical origin tracing approaches for teas

Methods

Results

Conclusion