Premium
DESCRIPTIVE ANALYSIS OF TOASTED ALMONDS: A COMPARISON BETWEEN EXPERT AND SEMI‐TRAINED ASSESSORS
Author(s) -
GUERRERO LLUIS,
GOU PERE,
ARNAU JACINT
Publication year - 1997
Publication title -
journal of sensory studies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.61
H-Index - 53
eISSN - 1745-459X
pISSN - 0887-8250
DOI - 10.1111/j.1745-459x.1997.tb00052.x
Subject(s) - profiling (computer programming) , descriptive statistics , computer science , sensory analysis , vocabulary , natural language processing , artificial intelligence , mathematics , machine learning , statistics , linguistics , philosophy , operating system
Classical techniques of descriptive sensory analysis are useful in food characterization. However, these methods need trained assessors and an agreed vocabulary. Free Choice Profiling may be regarded as a good alternative as it allows descriptive profiles to be carried out with untrained assessors, although certain problems may arise in the generation of descriptors and their interpretation. The descriptive profiles, regarding different lots of toasted almonds, generated by a panel of semi‐trained assessors and by a panel of expert assessors were compared. The resulting descriptors were basically the same for both panels. The semi‐trained assessors generated less specific and, sometimes, ambiguous and redundant descriptors. The expert assessors showed a higher efficiency in the generation of descriptors, especially regarding texture. Both profiles were graphically similar, though a deeper analysis showed that only the expert assessors were able to discriminate very similar products. Using a panel of semi‐trained assessors solves the main problems which arise when consumers are used in free choice profiling. However, both panels are not capable of discriminating very similar products.