Rapid determination of the geographical origin of Chinese red peppers (Zanthoxylum Bungeanum Maxim.) Based on Sensory Characteristics and Chemometric Techniques

Abstract

In this paper, principal component analysis (PCA), linear discriminant analysis (LDAp, artificial neural networks (ANN), and support vector machine (SVM) were applied to discriminate the geographical origin of Chinese red peppers (Zanthoxylum bungeanum Maxim.). The models based on color, smell and taste may discriminate quickly and effectively the geographical origin of Chinese red peppers from different regions, but the successful identification rates may vary with different kinds of parameters and chemometric methods. Among them, all models based on taste indexes showed an excellent ability to discriminate the geographical origin of Chinese red peppers with correct classifications of 100% for the training set and the 100% for test set. The present study provided a simple, efficient, inexpensive, practical and fast method to discriminate the geographical origin of Chinese red peppers from different regions, which was of great importance for both consumers and producers.