Premium
Chemometric classification of two peach palm ( Bactris gasipaes H.B.K.) landraces (Juruá and Vaupes)
Author(s) -
Rojas J. M.,
Serruya H.,
Bentes M. H. S.
Publication year - 1994
Publication title -
journal of the american oil chemists' society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.512
H-Index - 117
eISSN - 1558-9331
pISSN - 0003-021X
DOI - 10.1007/bf02541546
Subject(s) - principal component analysis , palm , mathematics , pattern recognition (psychology) , data set , palm oil , artificial intelligence , horticulture , botany , statistics , biology , computer science , food science , physics , quantum mechanics
A multivariate analysis by Principal Components (PCA), Hierarchical Cluster Analysis, K‐Nearest Neighbor (KNN) and Soft Independent Modeling of Class Analogy (SIMCA) methods were used to classify plants to different peach palm races. These statistical operations were applied to a data set of nineteen peach palm plant samples. Each data set contained fifteen variables defined as chemical characteristics of the mesocarp flour and physicochemical characteristics of the oil. The plants belonged to two different races. PCA showed that two principal components separated these races into two classes. KNN and SIMCA confirmed this classification. The final data for the model contained sixteen samples (plants) and eight variables. These results showed the utility of using chemometric methods for the classification of botanical species. These methods should aid the identification of new sources of oleaginous plants.