Two-stage consensus model based on opinion dynamics and evolution of social power in large-scale group decision making

Abstract

The main challenges in large scale group decision making (LSGDM) problem are how to tackle with the great number of participants and how to achieve a common solution accepted by most of participants. To address these challenges, in this paper, we propose a novel framework based on opinion evolution to study the consensus reaching process (CRP) in the LSGDM. In the proposed framework, we focus on the CRP in a dynamical social influence relationship context and the whole CRP is divided into two stages. In the first stage, we design a social power and opinion evolution iterative algorithm to estimate the final consensus opinions in each sub-group. In the second stage, the opinion leaders are selected as the representatives of each sub-group to participate in the consensus process. Subsequently, we develop a self-appraisal mechanism to evaluate the confidence degree of each opinion leader. Furthermore, an opinion leaders’ networked preference evolution mechanism is proposed to investigate the consensus formation of the second stage. Finally, we use this model on a case study and compare it with some existing models.