Large-scale Group Consensus Decision-making Method Based on Overlapping Community Division

Group decision-making method based on social network can make full use of the social network relationship between members to promote group consensus. This study considers the relationship network among group members and proposes a large-scale group consensus decision-making method based on overlapping community division algorithm. First, an overlapping community detection algorithm based on the tag propagation is employed to classify large number of members into several overlapping subgroups. Second, a weight determination method based on social network analysis and members’ preferences is used to derive the weights of individuals and subgroups, respectively. Then, several preference adjustment strategies considering different consensus levels are designed; a behavior management method based on the Uninorm aggregation operator is established to address non-cooperative behaviors; and an opinion substitution strategy is involved to copy with the preferences of individual members who are reluctant to revise. With the help of social network analysis method, the proposed large-scale group consensus method further explores the problem of individual weight distribution in the overlapping community environment and various behavior management methods. The effectiveness of the proposed method is verified by illustrative examples and simulation experiments.