Classification of rice wine according to different marked ages using a portable multi-electrode electronic tongue coupled with multivariate analysis

A developed multi-electrode electronic tongue combined with multivariate analysis was attempted to classify Chinese rice wine according to four different marked ages (3, 5, 8 and 10 years) for authentication. The electronic tongue consisted of three working electrodes (glassy carbon, gold and platinum) in a standard three-electrode configuration, using cyclic voltammetry technique to record signals. Characteristic values were extracted from the raw data signals for further multivariate analysis. In developing the discrimination models, linear (SIMCA, PLSDA and KNN) and nonlinear (BPANN and SVM) pattern recognition methods were comparatively employed, and they were optimized by cross-validation. Compared with other models, BPANN model achieved the best result, with the identification rate of 100% in the calibration set, and 100% in the prediction set. The overall results showed that this portable multi-electrode electronic tongue system with BPANN classification tool could successfully be used in identification of rice wine with different marked ages. (C) 2012 Elsevier Ltd. All rights reserved.