APPLICATION OF MULTIVARIATE STATISTICS FOR THE QUALITY CLASSIFICATION OF SINGLE-STRENGTH GRAPEFRUIT JUICE

Grapefruit juice volatiles were collected and analyzed by gas-liquid chromatography for 24 samples of commercial single-strength grapefruit juice from different production days and storage conditions. With the help of multivariate statistics, attempts have been made to classify them according to the chromatographic data and sensory evaluation of the juice. The data was analyzed by canonical and cluster analysis. As a typical result, it was found that using three constituents, the samples could be classified into four different quality classes. From the point of view of sensory interpretation, the canonical analysis also proved to be an interesting approach. Of the three more significant constituents, nootkatone and an unknown constituent have positive contribution, while another unknown constituent give a reverse contribution.