34 / 2017-05-10 23:14:37

Discrimination of traceability indexes of rice based on multi-element fingerprinting via pattern recognition techniques

rice, traceability, multi-element, principal component analysis, cluster analysis, discriminant analysis

The multi-element contents of 11 parameters include: Cd、Pb、Cr、Cu、Fe、Zn、K、Mn、Na、Mg、Ca of 100 rice samples from 10 origins in Jilin Songyuan were determined by using the Atomic Absorption Spectroscopy (AAS). To investigate the significance of the multi-element on tracing the geographical origin of rice, and to seek the multi-element indexes to characterize the region characteristics, principal component analysis(PCA), cluster analysis(CA) and discriminant analysis(DA) have been used. From 11 elements measured, 6 were selected as the first principal component is mainly composed of Zn, K, Mg, Na, Ca and Mn by PCA, except Pb, the results were consistent with the CA, and the contribution rate of the first principal component is the largest, accounting for 47.176%. The application of discriminant analysis gave correct assignation percentages of 100% for the multi-element indexes have been selected.