Abstract: Different beer brands and their mixtures were tested by electronic tongue and the response signals were analyzed by principal component analysis (PCA), discriminant factor analysis (DFA) and partial least-squares analysis (PLS). Individual beer brands and their mixtures were sucessfully recognized by electronic tongue. Moreover, a high regression coefficient, R2 = 0.9436, between instrumental response signal and mixed ratio of Snow,, beer and Blue Sword,, beer was found for the established PLS model with a prediction error ranging from 1.43% to 3.00%. This study demonstrates that electronic tongue is applicable for beer discrimination.

Key words: electronic tongue, principal component analysis (PCA), discriminant factor analysis (DFA), partial least-squares analysis (PLS), beer