Integration of Dependent Features on Sensory Evaluation Processes

• Published in: Rough Sets and Current Trends in Computing pp. 240 - 249
• Main Authors: Espinilla, Macarena, Martínez, Francisco J., Estrella Liébana, Francisco Javier
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing 2014
• Series: Lecture Notes in Computer Science
• Subjects: decision analysis; dependence; interaction; linguistic information; Sensory evaluation; sensory information
• ISBN: 331908643X; 9783319086439
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-319-08644-6_25

The aim of a sensory evaluation process is to compute the global value of each evaluated product by means of an evaluator set, according to a set of sensory features. Several sensory evaluation models have been proposed which use classical aggregation operators to summary the sensory information, assuming independent sensory features, i.e, there is not interaction among them. However, the sensory information is perceived by the set of human senses and, depending on the evaluated product, its sensory features may be dependent and present interaction among them. In this contribution, we present the integration of dependent sensory features in sensory evaluation processes. To do so, we propose the use of the fuzzy measure in conjunction with the Choquet integral to deal with this dependence, extending a sensory evaluation model proposed in the literature. This sensory evaluation model has the advantage that offers linguistic terms to handle the uncertainty and imprecision involved in evaluation sensory processes. Finally, an illustrative example of a sensory evaluation process with dependent sensory features is shown.