Large-scale group classification decision making method and its application with trust–interest dual factors in social network

Abstract

The wide application of computer technology promotes the development of large-scale group decision making. However, with the increase of the number of people involved in decision-making, the complexity of group decision making process is also expanded. In this paper, we propose a large-scale group classification decision making method with trust–interest dual factors in social network, which reduce the dimension of large group based on the deep mining of social network between decision makers, and aggregate the fuzzy decision information in the form of cloud model to get the decision results. Specifically, we define the elastic modularity and the utility function to describe the trust-characteristic and interest-characteristic of subgroups, divide the experts into high-trust communities and high-interest coalitions with the corresponding algorithm respectively. We set the preference of expert in subgroup as cloud droplet, and the preference of the subgroup is gathered in the form of cloud model. The merged subgroup of isolated decision maker is determined according to the preference membership matrix between the preference of isolated decision maker and each subgroup. The feasibility of the method is verified by the combinatorial technology selection for ultra-low emission transformation of large coal-fired power plants.