Affiliations: [a] The School of Computer Science and Technology, Xi’an University of Posts and Telecommunications, Xían, China
| [b] Xi’an Key Laboratory of Big Data and Intelligent Computing, Xían, China
| [c] The Shanxi Key Laboratory of Network Data Analysis and Intelligent Processing, Xían, China
| [d] Department of Computer Science, University of Huddersfield, Huddersfield, United Kingdom
Abstract: Methods on the basis of the consensus reaching process are prevalent in Group Decision Making (GDM), which typically forces some evaluators to revise initial opinions in order to reach group consensus without being able to precisely reflect original viewpoints. Furthermore, in case the correct opinion is embedded in the hand of the minority, existing methods may not reach the correct consensus. With the aim to tackle these observations, a novel approach of the Positive and Negative Prediction Selection Rate (PNPSR) is proposed on the basis of the Pythagorean Fuzzy Preference Relation (PFPR) which enables to present individuals’ opinions in a pairwise manner using the linguistic preference relation. The PFPR expressed opinions then serve as input for the computation of the proposed PNPSR, the minimum of which is subsequently selected as the correct one. Finally, the full ranking of the alternatives can be calculated through the proposed iterative algorithm. In the process, the evaluators’ original opinions are not required to modify, and the correct result can be achieved when the minority evaluators provide the correct opinions. Experimental results demonstrate the efficacy of the proposed approach in comparison with two state-of-the-art methods.
Keywords: Group decision making, Pythagorean fuzzy preference relation, positive and
negative prediction selection rate, consensus measure, consensus reaching process
