Predicting Bordeaux red wine origins and vintages from raw gas chromatograms

Authenticity Markers of Aged Red Wines from Aglianico, Uva di Troia, Negroamaro and Primitivo Grapes

The wide ampelographic treasure of Italian wine grape varieties is driving research towards suitable approaches for the varietal authenticity control of wine. In this paper, Aglianico, Negroamaro, Primitivo and Uva di Troia red wines, which were produced experimentally by single-grape winemaking from non-aromatic grapes native to southern Italy, were analyzed with respect to berry markers, namely anthocyanins, hydroxycinnamic acids (HPLC-DAD), shikimic acid (HPLC-UV) and glycosidic aroma precursors (GC-MS). The study confirms that, just as for the berries, useful varietal authenticity markers for red wine, even after aging, turn out to be hydroxycinnamic acids, relative amounts of acylated forms of anthocyanins, and shikimic acid, together with some grape glycosidic precursors from terpenes and C13- norisoprenoids. Principal Component Analysis was used as a valuable tool to highlight the results.

Deep reinforcement learning classification of sparkling wines based on ICP-MS and DOSY NMR spectra

An approach that combines NMR spectroscopy and inductively coupled plasma mass spectrometry (ICP-MS) and advanced tensor decomposition algorithms with state-of-the-art deep learning procedures was applied for the classification of Croatian continental sparkling wines by their geographical origin. It has been demonstrated that complex high-dimensional NMR or ICP-MS data cannot be classified by higher-order tensor decomposition alone. Extension of the procedure by deep reinforcement learning resulted in an exquisite neural network predictive model for the classification of sparkling wines according to their geographical origin. A network trained on half of the sample set was able to classify even 94% of all samples. The model can particularly be useful in cases where the number of samples is limited and when simpler statistical methods fail to produce reliable data. The model can further be exploited for the identification and differentiation of sparkling wines including a high potential for authenticity or quality control.