Uncorking White Wine Liking: Analytical Chemistry and Consumer Ratings

Consumer wine ratings can reflect both chemical composition and information such as price, labels, and region. These data come from a study focused on German white wines and support research into how measurable chemistry relates to consumer liking by combining laboratory measurements with crowd-sourced Vivino ratings. Two complementary types of laboratory data describe the wines. Gas chromatography-mass spectrometry (GC-MS) provides 145 measurements of volatile compounds related to aroma. Fourier-transform infrared spectroscopy (FTIR) provides 18 measurements of basic composition, including alcohol, acidity, pH, organic acids, glycerol, density, and reducing sugar. Supplementary material.xlsx is organized into four tables: sample descriptions, growing area, grape variety or wine type, vintage, and Vivino ratings; relative peak areas for 145 volatile organic compounds; 18 FTIR physicochemical measurements; compound names, IUPAC names, and chemical classes. Vivino ratings are available for 63 wines. The archive also contains 90 raw GC-MS chromatograms in NetCDF/CDF format for samples W001-W090. The combined data support exploration of relationships between aroma compounds, basic wine composition, and consumer liking, as well as multiblock analysis, feature selection, and quality-prediction research. Original dataset page: https://ucphchemometrics.com/uncorking-wine-quality-predicting-consumer-ratings-from-analytical-chemistry/ Original ERDA package: https://sid.erda.dk/share_redirect/GmGlS705zP/Vivino_GC_IR.zip If you use this data, please cite:Hjertholm, F.; Goetz, R.; Schneide, P.-A. A.; Petersen, M. A.; Bro, R.; Quintanilla-Casas, B. (2025). Uncorking white wine liking: Combining analytical chemistry and chemometrics with crowd-sourced data to predict quality ratings. Food Chemistry, 492, 145376. https://doi.org/10.1016/j.foodchem.2025.145376