Premium
EVALUATING CONSUMER ACCEPTANCE TESTS BY THREE‐WAY INTERNAL PREFERENCE MAPPING OBTAINED BY PARALLEL FACTOR ANALYSIS (PARAFAC)
Author(s) -
NUNES CLEITON A.,
PINHEIRO ANA CARLA M.,
BASTOS SABRINA C.
Publication year - 2011
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.2011.00333.x
Subject(s) - interpretability , preference , computer science , selection (genetic algorithm) , principal component analysis , data mining , artificial intelligence , statistics , mathematics
In this work, it is demonstrated that consumer acceptance analysis can be evaluated by simultaneously considering several attributes using a three‐way internal preference map obtained by parallel factor analysis (PARAFAC). Considerations regarding the building of this three‐way map by PARAFAC are reported. Pilot case studies with real data sets from herb cakes and beef burgers are also carried out, and comparisons with results from regular internal preference maps are obtained by principal component analysis. Three‐way internal preference maps enable the simultaneous analysis of interactions among consumer preferences, products and different evaluated attributes, which facilitate the selection of favorite samples. This method highlights the efficiency of the three‐way analysis of consumer acceptance data with different sources of data variability, allowing the extraction of relevant information and the graphic display of this information with improved interpretability. Three‐way internal preference mapping is a useful tool for the analysis of consumer acceptance tests, which can provide a more evidence‐based and general interpretation of data. PRACTICAL APPLICATIONS Three‐way internal preference mapping is another useful tool for the analysis of consumer acceptance tests, allowing the extraction of more relevant information and the graphic display of this information with improved interpretability. This tool makes it possible to simultaneously analyze the interactions among consumer preferences, products and different evaluated attributes, which can facilitate the selection of favorite samples. Furthermore, it enables a comparison of the overall performance of the samples in consumer acceptance tests, simultaneously taking into account the influence of all analyzed attributes. This method is useful in new product development and product improvement studies in research institutions and industries.