Potentiometry lipid membrane based electronic tongue for the classification of mint in tea by principal component analysis (PCA) and linear discrimination analysis (LDA)

Abstract

Abstract The principles of the electronic tongue sensor include potentiometry using a type of electrode with a liquid junction design. Mint leaves are widely used in the industrial sector in food, beverage, and medicinal products. In the industrial sector there are obstacles in the quality control process when analyzing the content in beverage products, in which the tools are unable to accurately classify the ingredients in the product. The content of a mixture of mint and tea was tested using seven working electrodes coated with different lipid membranes and classified using principal component analysis (PCA) and linear discriminant analysis (LDA). The electronic tongue was tested on a solution representing the five basic tastes. The accuracy of the readings of each sensor was obtained using a support vector regression (SVR) linear model with the mean absolute error (MAE) equal to 0.43 and a correlation coefficient ( ) of 0.93.