Journal of Chromatography A, 1216 (2009) 1458–1462 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Traceability of honey origin based on volatiles pattern processing by artiﬁcial neural networks  Tomas Cajka, Jana Hajslova ∗ , Frantisek Pudil, Katerina Riddellova Institute of Chemical Technology, Prague, Faculty of Food and Biochemical Technology, Department of Food Chemistry and Analysis, Technicka 5, 166 28 Prague 6, Czech Republic article info Article history: Received 28 August 2008 Received in revised form 1 December 2008 Accepted 22 December 2008 Available online 27 December 2008 Keywords: Honey Traceability Origin Authenticity Head-space solid-phase microextraction Comprehensive two-dimensional gas chromatography Time-of-ﬂight mass spectrometry 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 chemo- metric 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. 1. Introduction The European Union legislation (2001/110/EC) deﬁnes honey as “the natural sweet substance produced by Apis mellifera bees from the nectar of plants or from secretions of living parts of plants or excretions of plant-sucking insects on the living parts of plants, which the bees collect, transform by combining with speciﬁc substances of their own, deposit, dehydrate, store and leave in hon- eycombs to ripen and mature” [1]. Besides water, honey consists mainly of the monosaccharides (fructose and glucose) and many other substances such as organic acids, oligosaccharides, enzymes, vitamins, minerals, pigments, a wide range of aroma compounds, and solid particles derived from honey collection are present [2]. Honey is popular not only as a source of energy but also for its potentially health-promoting properties provided by prebiotic, antioxidant, antibacterial, and/or antimutagenic functionalities of certain constituents [3–6]. The price of honey is usually dictated by its botanical and/or geographical origin. While in the case of  The information reported reﬂects the authors’ views; the European Commission is not liable for any use of the information contained therein. Mention of brand or ﬁrm names in this publication is solely for the purpose of providing speciﬁc infor- mation and does not imply recommendation or endorsement by the Institute of Chemical Technology, Prague. ∗ Corresponding author. Tel.: +420 220 443 185; fax: +420 220 443 184. E-mail address: jana.hajslova@vscht.cz (J. Hajslova). botanical origin the most expensive are uniﬂoral honeys, in the later case the higher price arises when honey is produced in a speciﬁc geographic location. Up to now, the EU has speciﬁed 18 protected denomination of origin (PDO) regions for honey (one Greek, one Italian, one Luxemburgian, one Polish, two French (including the island of Corsica), three Spanish, and nine Portuguese) [7]. Recently, an increased number of alerts concerning safety (presence of a vari- ety of unauthorised or prohibited antimicrobial substances) and adulteration of honey have been posted [8–10]. In general, the adul- teration techniques of honey are based on various principles: (i) water addition and extension with sugar and/or syrups; (ii) bee feeding with sugars and/or syrups or artiﬁcial honey; (iii) misla- belling as a results of mixing of honeys originating from different ﬂoral or geographical origin [2]. For the honey characterisation various parameters such as pollen analysis, moisture content, 5-(hydroxymethyl)furan-2- carbaldehyde concentration, sugar composition, proline content, invertase and diastase activity, and electrical conductivity are typi- cally considered [2,11]. In addition to these traditional approaches, examination of the volatiles proﬁle might be considered as a strat- egy enabling honey authentication since its composition (volatiles including) is known to vary widely with the ﬂoral origin and way of processing [12]. During recent years, solid-phase microextraction (SPME) in combination with the gas chromatographic–mass spectromet- ric (GC–MS) technique has been implemented as a method of 0021-9673/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2008.12.066