RT Journal Article
SR Electronic
T1 Classification of Spanish Red Wines by Using Artificial Neural Networks with Oenological Parameters and Mineral Content
JF American Journal of Enology and Viticulture
JO Am J Enol Vitic.
FD American Society for Enology and Viticulture
SP ajev.2017.17021
DO 10.5344/ajev.2017.17021
A1 Isabel M. Moreno
A1 Angel J. Gutiérrez
A1 Carmen Rubio
A1 A. Gustavo González
A1 Dailos Gonzalez-Weller
A1 Naouel Bencharki
A1 Arturo Hardisson
A1 Consuelo Revert
YR 2018
UL http://www.ajevonline.org/content/early/2018/01/04/ajev.2017.17021.abstract
AB Spanish red wines from the Canary Islands were categorized into seven classes, and 20 samples from each DO studied, Tacoronte-Acentejo (class T), Valle de la Orotava (class O), “Ycoden-Daute-Isora” (class YDI), Abona (class A), Valle de Güímar (class VG) “La Gomera” (class G) and “La Palma” (class P) were studied. Metals (B, Ca, Cd, Co, Cr, Cu, Fe, K, Li, Mg, Mn, Mo, Ni, Na, Pb, and Zn) and physico-chemical (pH, volatile acidity, total acidity, malic acid, acetic acid, reducing sugars, alcohol content, free sulphur dioxide, total sulphur dioxide, and total polyphenols) were considered as suitable descriptors to differentiate between classes. Supervised learning pattern recognition procedures were applied. Linear discriminant analysis led to good results up to about 80% of correct classification. To improve the results, another kind of algorithms able to model non-linear separation between classes was considered: Artificial Neural Networks. Accordingly, excellent results were obtained, with 100% of the 140 wines correctly placed in the associated seven classes. Our results are in good agreement with the working hypothesis of differentiation between wines coming from different locations including different islands and also different sites in the Tenerife Island.