Fuzzy discriminant analysis of medicinal plant extracts according to their total content of phytochemicals and the antioxidant capacity

Fuzzy linear discriminant analysis, a robust supervised method, has been successfully applied for characterization and classification of 42 Romanian medicinal plant extracts according to the total content of eight phytochemical compounds (alkaloids, polyphenols, coumarins, o-diphenols, flavonoids, anthocyanins, flavonols, flavanols) estimated by dedicated molecular absorption spectrophotometry-based methods, and their antioxidant capacity determined by DPPH* method. The obtained results (fuzzy partitions) and parameters of the class centers (robust fuzzy means) clearly demonstrated the efficiency and information power of the advanced fuzzy method in plants characterization and classification, and allow a rationale choice of a medicinal plant extract with a specified phytochemical composition and/or antioxidant activity. Previous studies have investigated the association of certain classes of phytochemicals with the antioxidant activity in plant extracts. However, most of them are limited either in the number of the plants extracts they have analyzed or in the number of the employed phytochemical classes. The distinctness of this work is the application of a fuzzy multivariate analysis on data obtained for high number of plant extracts-42 widespread medicinal plants form various plant taxa-and a high number of ubiquitous phytochemical classes in plants. The methodology developed in this paper might be also extended in the authenticity and origin control of other fruits, herbs or derived products.