Characterization and Classification of Ohio Wines Using Multivariate Data Analysis

• Autores: James K. Hardy[sup*]
• Localización: Journal of wine research, ISSN 0957-1264, Vol. 10, Nº 3, 1999, págs. 197-205
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Multivariate data analysis was conducted on a series of 41 wine samples, representing two grape varieties, Chardonnay and Johannisberg Riesling, obtained from both Ohio and California. Each was assayed for both organic and metal content and the results used to construct a data matrix. The multivariate statistical methods of principal component analysis and K‐nearest neighbour were evaluated to determine if they would be successfully applied to classify samples based on both region and wine type. Classification results of wine samples displayed significant differences among regions and types with wine type providing the greatest source of variation.