Combining hedonic information and CATA description for consumer segmentation

Check-all-that-apply (CATA) has become a popular method for obtaining consumer-based sensory characterization. In most case studies, consumers are also asked to evaluate the set of products according liking scale with aim identify key attributes associated liked, or disliked, products. The common approach consists, first, in identification consumer segments based on preference profiles. Thereafter, analysis CATA responses is performed within each segment. Our purpose herein investigate different ways simultaneously clusters profiles while taking into account attributes. These approaches derived from strategies already proposed by co-authors, namely: Fuzzy Clusterwise Regression (FCR), Clustering around Latent Variables (CLV) external data, CLUSCATA-liking and CLV3W. first two involve aggregation individual data contingency table, last ones deal combination at level. four illustrated basis real study. Results compared respect cluster stability together interpretability results, assessed bootstrapping, differed strategy used. Moreover, working level combined lead somewhat segmentation panel consumers.