Abstract
In the last decade, temporal dominance of sensations (TDS) methods have proven to be potent approaches in the field of food sciences. Accordingly, thus far, methods for analyzing TDS curves, which are the major outputs of TDS methods, have been developed. This study proposes a method of bootstrap resampling for TDS tasks. The proposed method enables the production of random TDS curves to estimate the uncertainties, that is, the 95% confidence interval and standard error of the curves. Based on Monte Carlo simulation studies, the estimated uncertainties are considered valid and match those estimated by approximated normal distributions with the number of independent TDS tasks or samples being 50–100 or greater. The proposed resampling method enables researchers to apply statistical analyses and machine-learning approaches that require a large sample size of TDS curves.
Highlights
Over the last decade, temporal dominance of sensations (TDS) methods, in which multiple types of temporally evolving subjective responses are recorded, have been proven to be effective methods of sensory appraisal in the field of food sciences by many researchers [1,2,3]
Following the principles of statistical estimation, a greater sample size leads to smaller confidence intervals
The TDS method is a time-series sensory appraisal method that has been increasingly used in the field of food science
Summary
Temporal dominance of sensations (TDS) methods, in which multiple types of temporally evolving subjective responses are recorded, have been proven to be effective methods of sensory appraisal in the field of food sciences by many researchers [1,2,3]. In the task of the TDS method, adjective descriptors (Except for adjective descriptors, onomatopoeic words are used, for example [8]) listed on a computer screen are sequentially selected by assessors These descriptors correspond to the dominant sensations felt while experiencing food stimuli. As the most typical analysis [9], by accumulating the results of independent TDS tasks, a proportion that a certain descriptor is dominant in is computed at each instant. These proportions are functions of continuous time and are displayed as “TDS curves.”. These proportions are functions of continuous time and are displayed as “TDS curves.” These TDS curves are visually inspected, and their key values, such as maximum values and the time when the maximum values are observed, are calculated and compared using multivariate analysis techniques among different food products [2,14,15]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have