Premium
USE OF CONFIDENCE ELLIPSES IN A PCA APPLIED TO SENSORY ANALYSIS APPLICATION TO THE COMPARISON OF MONOVARIETAL CIDERS
Author(s) -
HUSSON FRANÇOIS,
BOCQUET VALÉRY,
PAGÈS JÉRÔME
Publication year - 2004
Publication title -
journal of sensory studies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.61
H-Index - 53
eISSN - 1745-459X
pISSN - 0887-8250
DOI - 10.1111/j.1745-459x.2004.062104.x
Subject(s) - principal component analysis , ellipse , mathematics , pattern recognition (psychology) , confidence interval , artificial intelligence , statistics , resampling , food science , computer science , biology , geometry
In a descriptive sensory analysis, a principal component analysis (PCA) is usually applied to analyze products with attributes. To visualize the variability of a mean point (i.e., the center of gravity of the product), we introduced confidence ellipses. In this particular application, resampling methods are performed. Twenty products composed of five varieties of cider were studied. We would like to know if these five varieties are different. Confidence ellipses reveal differences between varieties foreshadowed in PCA.