Application of emission - excitation matrices in parallel with factor analysis with other chemometric techniques for honey classification

Fluorescence spectroscopy in conjunction with parallel factor analysis (PARAFAC), principal component analysis (PCA) and SIMCA was used for the development of geographic and botanical discrimination models to differentiate among distinct honey classes. For this aim, 96 authentic honey samples, having several botanical origins (acacia, chestnut, colza, honeydew, lavender, linden and sunflower), originated from Romania and France, were involved in this study. Excitation emission spectra were obtained for all investigated samples by recording the emission from 275 to 600 nm with excitation in the range of 250-500 nm. By using this approach, a classification rate of 95.8 % was obtained for the geographical differentiation of the samples while, for the simultaneous differentiation of the four predominant honey varieties, in terms of representatively per class (acacia, colza, linden and sunflower), a correct percentage of 94.5 % was achieved.

Automated summary

This study developed geographic and botanical discrimination models for differentiating honey classes using fluorescence spectroscopy combined with parallel factor analysis, principal component analysis, and SIMCA. The classification rate for geographical differentiation of the samples was 95.8%, and a correct percentage of 94.5% was achieved for differentiating the four predominant honey varieties.