Multi-classification Decision-Making Method for Interval-Valued Intuitionistic Fuzzy Three-Way Decisions and Its Application in the Group Decision-Making
Sichuan College of Architectural Technology,Liang Decui,Li Tao,Liang Shujing
DOI: https://doi.org/10.1007/s13042-020-01195-3
2020-01-01
International Journal of Machine Learning and Cybernetics
Abstract:With the introduction of the interval-valued intuitionistic fuzzy sets, the interval-valued intuitionistic fuzzy numbers are used instead of precise numbers to provide fuzzy characterization of feature attribute values and misclassification loss function values, which is more in line with the realistic fuzzy decision-making environment. Also, the constructive covering algorithm is introduced into the three-way decisions model, which effectively solves the shortcomings of the traditional decision-theoretic rough sets model in dealing with multi-classification problems, such as too many artificial parameters, complicated computation, redundant decisions, decisional conflicts and excessively large boundary domains. At the same time, in order to avoid the one-sidedness of individual decisions, the group decision-making method is introduced into the preliminarily constructed multi-classification model in this paper to build a multi-classification group decision-making model for interval-valued intuitionistic fuzzy three-way decisions based on the constructive covering algorithm. This model determines the initial weights of feature attributes by the precise weighting method, and determines the expert weights by the grey relational precise weighting method, which effectively achieves the consistency of group decision-making. The decision-making process and rules are also deduced, which expand the model of three-way decisions as well as its practical application value and scope.