Journal : Food Quality and Preference , vol. 15 , p. 681–687–7 , 2004
Publisher : Elsevier
International Standard Numbers
Printed : 0950-3293
Electronic : 1873-6343
Publication type : Academic article
Issue : 07.aug
If you have questions about the publication, you may contact Nofima’s Chief Librarian.
This paper presents various strategies for the identification of consumer segments. One issue is to employ some segmentation method on the “long-thin” versus “short-fat” orientation of the consumer preference data. One strategy is to assign consumers with no strong preference to one segment, and then apply fuzzy clustering (FC) in the segmentation on the remaining consumers. The segments are evaluated by ANOVA and from external preference mapping to sensory descriptive data. Shuffled preference data are subject to segmentation to investigate the robustness of the procedure. The results indicate that segmentation on the long-thin data gives a better relation to the sensory data, and that segmentation on the short-fat data without defining the non-reliable segment gives rise to spurious segments.