Hierarchical Clustering Analysis of Temporal Dominance of Sensations Tasks

Abstract

Temporal dominance of sensation (TDS) is used to obtain the time-series of multiple types of sensory responses during food consumption. Results of existing analysis methods for TDS are the averages of all panels. Individual differences and their time-series similarity have seldom been studied. We defined a dissimilarity index to compare individual TDS trials and performed hierarchical clustering on TDS data of strawberries using the index. Fifty-one trials were classified into four clusters with different characteristics. The method will be useful for individual analyses of time-series sensory evaluation data.