ORANGE JUICE CLASSIFICATION WITH A BIOLOGICALLY-BASED NEURAL-NETWORK

Dystal, an artificial neural network, was used to classify orange juic e products. Nine varieties of oranges collected from six geographical regions were processed into single-strength, reconstituted or frozen c oncentrated orange juice. The data set represented 240 authentic and 1 73 adulterated samples of juices; 16 variables [8 flavone and flavanon e glycoside concentrations measured by high-performance liquid chromat ography (HPLC) and 8 trace element concentrations measured by inductiv ely coupled plasma spectroscopy] were selected to characterize each ju ice and were used as input to Dystal. Dystal correctly classified 89.8 % of the juices as authentic or adulterated. Classification performanc e increased monotonically as the percentage of pulpwash in the sample increased. Dystal correctly identified 92.5% of the juices by variety (Valencia vs non-Valencia).