Classification of Wines Based on Combination of ^1H NMR Spectroscopy and Principal Component Analysis

A combination of ^1H nuclear magnetic resonance （NMR） spectroscopy and principal component analysis （PCA） has shown the potential for being a useful method for classification of type, production origin or geographic origin of wines. In this preliminary study, twenty-one bottled wines were classified/separated for their location of production in Shacheng, Changli and Yantai, and the types of the blended, medium dry, dry white and dry red wines, using the NMR-PCA method. The wines were produced by three subsidiary companies of an enterprise according to the same national standard. The separation was believed to be mainly due to the fermentation process for different wines and environmental variations, such as local climate, soil, underground water, sunlight and rainfall. The major chemicals associated with the separation were identified.

Predictive modeling for wine authenticity using a machine learning approach

The purpose of this paper is to classify wines from 4 different countries in South America.Each class of wines is formed by samples considered by experts as representatives of the following commercial categories:“Argentinean Malbec(AM)”,“Brazilian Merlot(BM)”,“Uruguayan Tannat(UT)”and“Chilean Carménère(CC)”.The 83 samples collected were analyzed according to their composition of volatiles,semi-volatiles and phenolic compounds.We built a decision system for classification based on support vector machines(SVM),along with Correlation-based Feature selection(CFS),and RandomForest Importance(RFI),whichmeasures the relative importance of the input variables.First,we use CFS to select a subset of variables among 190 chemical compounds.Thirteen chemicals were selected as correlated to the category and uncorrelated with each other.Afterwards,these chemical compounds were organized according to the importance ranking given by the RFI and classified with SVM.The study clearly indicated that SVMin combination with feature selection methodswas able to identify the most important chemicals to classify the wine samples.Among the compounds identified in the wine samples,the variable subset defined by the feature selection methods,which were catechin,gallic,octanoic acid,myricetin,caffeic,isobutanol,resveratrol,kaempferol,and ORAC,were able to achieve an accuracy of 93.97%in classifying the commercial categories.