A1 Refereed original research article in a scientific journal
An approach for clustering consumers by their top-box and top-choice responses
Authors: Castura John C, Meyners Michael, Pohjanheimo Terhi, Varela Paula, Naes Tormod
Publisher: WILEY
Publication year: 2023
Journal: Journal of Sensory Studies
Journal name in source: JOURNAL OF SENSORY STUDIES
Journal acronym: J SENS STUD
Number of pages: 19
ISSN: 0887-8250
DOI: https://doi.org/10.1111/joss.12860
Web address : https://doi.org/10.1111/joss.12860
Abstract
Cluster analysis is often used to group consumers based on their hedonic responses to products. We give a motivating example in which conventional cluster analyses converge on a solution where consumers do not agree on which products they like. We show why this occurs. We state a goal: to group together consumers who have a shared opinion of which products are delightful and which products are not delightful, apart from consumers who have a different opinion. To meet this goal, we code consumers' hedonic responses in ways inspired by top-k box analysis, then cluster consumers using b-cluster analysis. For comparison, we cluster consumers using two conventional methods. We interpret each cluster by focusing on which product(s) the cluster accepts and whether a large proportion of cluster members are aligned in accepting these products. Solutions from b-cluster analysis based on top-k box-inspired codings met our goal better than conventional approaches, indicating that these methods deserve further study. Practical ApplicationsCluster analysis outcomes are profoundly shaped by a researcher's decisions related to response coding and clustering algorithm. This paper highlights the importance of determining the goal of the cluster analysis first, then selecting a response coding and clustering algorithm to best meet this goal. Our stated goal is one that is frequently of interest in sensory evaluation but is not well met by conventional clustering approaches. The novel approaches that we give in this paper meet the goal and are available using software that is freely available in the public domain.
Cluster analysis is often used to group consumers based on their hedonic responses to products. We give a motivating example in which conventional cluster analyses converge on a solution where consumers do not agree on which products they like. We show why this occurs. We state a goal: to group together consumers who have a shared opinion of which products are delightful and which products are not delightful, apart from consumers who have a different opinion. To meet this goal, we code consumers' hedonic responses in ways inspired by top-k box analysis, then cluster consumers using b-cluster analysis. For comparison, we cluster consumers using two conventional methods. We interpret each cluster by focusing on which product(s) the cluster accepts and whether a large proportion of cluster members are aligned in accepting these products. Solutions from b-cluster analysis based on top-k box-inspired codings met our goal better than conventional approaches, indicating that these methods deserve further study. Practical ApplicationsCluster analysis outcomes are profoundly shaped by a researcher's decisions related to response coding and clustering algorithm. This paper highlights the importance of determining the goal of the cluster analysis first, then selecting a response coding and clustering algorithm to best meet this goal. Our stated goal is one that is frequently of interest in sensory evaluation but is not well met by conventional clustering approaches. The novel approaches that we give in this paper meet the goal and are available using software that is freely available in the public domain.
UTU Research Portal