A k-core decomposition-based opinion leaders identifying method and clustering-based consensus model for large-scale group decision making

Abstract

Due to the development of network technology, large-scale group decision making (LSGDM) has become increasingly concerned. In this paper, a k-core decomposition-based opinion leaders identifying method and clustering-based consensus model are developed for LSGDM problems. Firstly, a clustering method based on similarity degree is provided for dividing decision makers (DMs) into several clusters. Then, sub-clusters are presented for social networks (SNs) construction process, which are consist of DMs with same alternative ranking information. Furthermore, a novel k-core decomposition-based opinion leaders identifying method is proposed for selecting opinion leaders of these SNs. Finally, the opinion leaders identified are applied to the following clustering-based consensus model in LSGDM. The weights of DMs are distributed appropriately and the group can efficiently reach a consensus based on the proposed social network analysis (SNA) methods and consensus reaching process (CRP). A case study on flood disaster management shows that the proposed methods are feasible for LSGDM problems.