View article

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 …