Affiliations: [a] School of Mathematics, Northwest University, Xi’an, Shaanxi, P.R. China | [b] School of Mathematics and Statistics, Chongqing University of Technology, Chongqing, P.R. China
Abstract: Recently, making decisions and analyzing data are getting more and more attention by taking advantage of rough set and intuitionistic fuzzy set theories. Additionally, it can be found that many works have been developed about intuitionistic fuzzy rough set approaches from different viewpoints. In this article, we introduce similarity degrees and four kinds of uncertainty measurement, called θ-conditional entropies, θ-similarity intuitionistic fuzzy accuracies, θ-similarity intuitionistic fuzzy roughness and θ-rough decision entropies in intuitionistic fuzzy decision tables. Also, we provide a novel method for classifying the objects’ intuitionistic fuzzy decision table. Moreover, we carefully discuss the lower approximation and upper approximation of a given set and classify their important properties based on the novel classes in the intuitionistic fuzzy decision table. Furthermore, an illustrated example is employed to demonstrate the conceptual arguments of these measurements based on different similarity degrees and similarity rates. From this, it can be found that the new measures are superior to the classical accuracy and roughness, and the method is valuable and useful in real life situations.
