Analyzing Sensory Data Using Non-linear Preference Learning with Feature Subset Selection

Abstract

The quality of food can be assessed from different points of view. In this paper, we deal with those aspects that can be appreciated through sensory impressions. When we are aiming to induce a function that maps object descriptions into ratings, we must consider that consumers' ratings are just a way to express their preferences about the products presented in the same testing session. Therefore, we postulate to learn from consumers' preference judgments instead of using an approach based on regression. This requires the use of special purpose kernels and feature subset selection methods. We illustrate the benefits of our approach in two families of real-world data bases.

Dimension Reduction for Supervised Ordering

“…In [17,2], supervised ordering methods are used for sensory tests to examine which product features affect the value of the products. Metasearch engines are constructed in [4,8].…”

“…In particular, several methods are being developed for learning functions used to sort objects represented by attribute vectors from example orders. We call this task Supervised Ordering [14] and emphasize its usefulness for sensory tests 1 [14,17], information retrieval [4,9,11,20,23] , and recommendation [8].…”

Analysis of nutrition data by means of a matrix factorization method

“…On the other hand, in [1,13,5,4], the sensitivity analysis of [15] was used to compute a ranking of features in a closely related context: learning preferences of users about a kind of items. Notice that its aim is also to learn to order things, and pairwise comparisons were also used to approach these learning tasks.…”

Prediction and Inheritance of Phenotypes

“…In [11,12] we proposed a new assessment method for beef cattle. On the other hand, in [13,14,15] we described a collection of methods to handle sensory data from consumer's opinions about food products.…”