Traceability of honey origin based on volatiles pattern processing by artificial neural networks

Abstract

Head-space solid-phase microextraction (HS-SPME)-based procedure, coupled to comprehensive two-dimensional gas chromatography-time-of-flight mass spectrometry (GC × GC-TOF-MS), was employed for fast characterisation of honey volatiles. In total, 374 samples were collected over two production seasons in Corsica (n = 219) and other European countries (n = 155) with the emphasis to confirm the authenticity of the honeys labelled as "Corsica" (protected denomination of origin region). For the chemometric analysis, artificial neural networks with multilayer perceptrons (ANN-MLP) were tested. The best prediction (94.5%) and classification (96.5%) abilities of the ANN-MLP model were obtained when the data from two honey harvests were aggregated in order to improve the model performance compared to separate year harvests. © 2008 Elsevier B.V. All rights reserved.