PROCESSOR, PROCESSING METHOD, PRESUMPTION DEVICE, PRESUMPTION METHOD, AND PROGRAM

摘要

PROBLEM TO BE SOLVED: To express a selection behavior of a perception-biased consumer, using a learnable model with high prediction accuracy, in view of feature quantities of a product and the consumer.SOLUTION: There are provided a processor which generates a selection model as a model of a selection behavior of a selecting subject selecting at least one from presented input options, a method for the processing, and a program. The processor includes: an acquisition part acquiring leaning data containing a plurality of input feature vectors representing a plurality of input options presented by the selecting subject and an output feature vector representing the feature of an output option, which is an option selected by the subject from the input options; an input synthesis part generating an input synthesis vector by synthesizing the input feature vectors; and a learning processing part leaning the selection model on the basis of the input synthesis vector and the output feature vector.SELECTED DRAWING: Figure 4