A personalized consensus-reaching method for large-group decision-making in social networks combining self-confidence and trust relationships

Abstract

In recent years, large-scale group decision-making (LSGDM) in social network environments considering experts’ psychological behaviors has received increasing attention. Moreover, existing studies have shown that whether it is internal self-confidence or external trust relationships of experts, they play a crucial role in reaching consensus. Therefore, this paper integrates self-confidence and trust relationships, and proposes a personalized consensus-reaching method for LSGDM from the perspective of adjustment willingness. Firstly, we explored the promoting effect of opinion similarity on the efficiency of trust propagation and proposed a method to evaluate unknown trust relationships among experts, integrating the objectivity of trust relationships and the subjectivity of self-confidence to determine the experts’ weights. Secondly, a hierarchical fuzzy clustering algorithm based on the chi-square test is proposed for effective subgroup division, which avoids the impact of setting initial clustering parameters on the clustering results. Afterwards, the adjustment willingness of the subgroups is determined by combining the experts’ self-confidence and the trust relationships between them. In addition to this, a personalized consensus feedback adjustment mechanism that synthesizes the adjustment willingness and trust relationship is constructed to reach consensus, which can better preserve the original information. Finally, the effectiveness of the proposed method is verified through a numerical example. In addition, the advantages of the proposed method are demonstrated by comparing with other methods.