Abstract : In this paper we propose to develop a method for learning strategies established by a decision maker for a feature task of the selection/elimination of objects described by several attributes (parameters ...). This method is based upon a cognitive model of the decision maker stemming from psychological research. We will explain its mathematical and algorithmical consequences. We will present an algorithm (and a way of coding the objects) which extracts rules used by the decision maker. The complexity and efficiency of the algorithm will be discussed. It will be illustrated with a concrete example and real data in the last section.
https://hal-imt-atlantique.archives-ouvertes.fr/hal-01893331 Contributor : Philippe LencaConnect in order to contact the contributor Submitted on : Friday, October 19, 2018 - 4:56:34 PM Last modification on : Thursday, April 22, 2021 - 3:33:08 AM Long-term archiving on: : Sunday, January 20, 2019 - 12:24:42 PM