Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10561578 | Talanta | 2005 | 6 Pages |
Abstract
Results were submitted to pattern recognition procedures, unsupervised methods such as cluster and principal components analysis and supervised learning methods like linear discriminant analysis in order to evaluate the existence of data patterns and the possibility of differentiation of Spanish honeys from different botanical origins according to their mineral content. Cluster analysis shows four clusters corresponding to the four botanical origins of honey and PCA explained 71% of the variance with the first two PC variables. The best-grouped honeys were those from heather; eucalyptus honeys formed a more dispersed group and finally orange-blossom and rosemary honeys formed a less distinguishable group.
Keywords
Related Topics
Physical Sciences and Engineering
Chemistry
Analytical Chemistry
Authors
Rut Fernández-Torres, Juan Luis Pérez-Bernal, Miguel Ángel Bello-López, Manuel Callejón-Mochón, Juan Carlos Jiménez-Sánchez, A. Guiraúm-Pérez,