Rapid screening of sensory attributes of mackerel using big data mining techniques and rapid sensory evaluation methods

The present study aimed to investigate the potential of big data mining technology in conjunction with rapid sensory evaluation methods for the swift screening of sensory attributes of three kinds of frozen mackerel. Specifically, two rapid sensory evaluation methods, namely ideal profile method (IPM) and check-all-that-apply (CATA), were implemented and compared with the conventional descriptive analysis method. The results revealed that eight sensory attributes based on consumer network evaluations demonstrated significant consistency during the training process (p < .05). Notably, the application of web-based sensory attributes yielded highly comparable results between IPM and traditional descriptive analysis (0.915). Moreover, the results of the IPM preference map were in closer agreement with those of traditional descriptive analysis. While traditional sensory evaluation boasts high accuracy and a greater ability to detect nuances, the evolution of sensory evaluation technology has shifted its focus toward consumers. Rapid sensory evaluation analysis technology supports the collection of information directly from consumers, even by untrained or semi-trained groups, thereby presenting broad prospects for product qualitative analysis.