An electronic tongue was used to classify honey samples of different floral origins and geographical origins. Eight kinds of honeys of different floral origins and five kinds of Acacia honeys of different geographical origins were detected. The data obtained were analyzed by three-pattern recognition techniques: Principal component analysis (PCA), Cluster analysis (CA), and Artificial neural network (ANN). It was possible to discriminate the eight kinds of honeys of different floral origins completely based on PCA, while good results were shown by CA and ANN, too; the five kinds of Acacia honey from different geographical origins could not be differentiated clearly by PCA, however, ANN was the most effective feature extraction method compared with CA and PCA, and the correction rate could reach to 95%. The soluble sugar content and conductance of the samples were also detected in this paper and some interesting regularity is shown in the score plots with the help of PCA. (C) 2009 Elsevier Ltd. All rights reserved.