A novel outcome evaluation model of three-way decision: A change viewpoint

Abstract

Three-way decision (3WD) is a granulation-computing paradigm consisting of trisecting, acting, and outcome (i.e., TAO model). Movement-based three-way decision (M-3WD) and change-based three-way decision (C-3WD) are two emerging studies of three-way decision. A central notion to studies of outcome measures is the change of trisection. In this paper, a novel outcome evaluation model of 3WD is established from a change viewpoint. The purpose of this model is to facilitate some ideal changes and prevent unideal changes. Moreover, the effectiveness measure of outcome aims to change, and it is also reflected in the change. Generally, three-way change (3WC) exists in the research results of many three-way decision models, i.e., Bayesian confirmation rough set, movement-based three-way decision, etc. Of course, three-way changes can also be regarded as a further extension and expansion of these models. With such ideas in mind, evaluation-based 3WC is presented by examining its basic ingredients and application. The effectiveness of the proposed model is verified by some numerical experiments carried out on public datasets. The considered framework can provide a convenient novel tool for the researches of 3WD.