Game-Theoretic Expert Importance Evaluation Model Guided by Cooperation Effects for Social Network Group Decision Making

The evaluation of expert importance degree for solving group decision-making problems (GDM) is meaningful, especially for social network GDM cases. Conventionally, the importance of experts in existing GDM models is assumed to be isolated. Nevertheless, in real-life scenarios, the internal components of expert systems should be mutually influential. In this study, a novel game-theoretic expert importance evaluation model guided by cooperation effects is proposed. First, the framework of non-additive fuzzy measure values is utilized to obtain the initial opinions of all experts. An interaction indicator is then exploited to represent peer interaction effort (PIE). With the log-sigmoid transition technique, individual social cooperation networks (ISCNs) are then constructed. With the advanced aggregation operator, the global social cooperation network (GSCN) of the corresponding expert collection can be generated. Eventually, a modified gravity model is designed to evaluate the degree of importance for the experts. Several experiments are conducted to demonstrate the effectiveness of the proposed method. The results show that the influence of cooperation effects can reasonably be considered in the expert importance evaluation procedure, which is beneficial to real-life scenarios. Additional comparisons and related discussions are also provided.