Outlier analysis in time-evolving sensory evaluation tasks

Abstract

The temporal dominance of sensations (TDS) method measures the time-evolving changes in subjectively reported sensations. This method has been used to obtain average experiences among panels, and there has been no method to analyze individual differences in such dynamic experiences. Therefore, we proposed a dissimilarity index to quantitatively judge how two TDS tasks deviate and then successfully screened an outlier among 51 TDS tasks for strawberries. The task judged as an outlier was semantically different from the average task.