Principal Component Analysis of Time Series Taste Data to Classify Processed Ham

抄録

Temporal dominance of sensations (TDS) is a sensory evaluation method that measures changes in several types of sensations over time during food sample tasting. Previous analytical methods for TDS do not classify different foods using the dynamicsensory information of the TDS method results. We developed a new approach to classify products by principal component analysis computation of the time-series information of sensory responses. This method could classify five different hams in a two-dimensionalprincipal component space. The time-series information possessed by each dimension was interpretable, which suggests that thedeveloped method is helpful for analyzing the temporal properties captured by TDS methods.