Wineinformatics: Can Wine Reviews in Bordeaux Reveal Wine Aging Capability?

Abstract

Wineinformatics is a new and emerging data science that uses wine as domain knowledge and integrates data systems and wine-related data sets. Wine reviews from Wine Spectator usually include the aging information, at the end of the review, in the form of “Best from YearA through YearB”; with the vintage of the wine included, the suggested holding year (YearA—vintage), shelf-life (YearB—vintage) and aging capacity (YearB—YearA) can be calculated and provide crucial information in the study of wineinformatics. The goal of this paper is to test whether wine reviews describing olfactory and gustatory information reveal wines’ suggested holding-year information. Wine reviews from Wine Spectator are extracted and processed by a natural language processing tool named the Computational Wine Wheel for categorizing and mapping various wine terminologies from wine reviews into a consolidated set of descriptors. The suggested aging capability is also calculated from the review and served as a label for classification problems. The study uses different learning algorithms, analyzing their performances and using the best-performing algorithm(s) to build a model for the prediction of a wine’s aging properties. The results of the study suggest that both support vector machine (SVM) and the K-nearest neighbor (KNN) algorithms achieved more than 70% accuracy. These results suggest that the algorithms are able of capturing a hidden informational relationship between a wine’s reviews and its aging capability.

Keywords

• wineinformatics
• data science
• machine learning
• Bordeaux
• aging capability