A Clustering Method with Historical Data to Support Large-Scale Consensus-Reaching Process in Group Decision-Making

Abstract With the rapid development of information technology and social network, large-scale group decision-making (LSGDM) has become more popular due to fact that large numbers stakeholders are involved in different decision problems. To support consensus-reaching process (LCRP), this paper proposes a LCRP framework based on clustering method with historical preference data all makers (DMs). There three parts proposed framework: process, consensus selection process. In we make use an extended k-means technique divide DMs into several clusters their preferences data. Next, consists measure feedback adjustment. The aims calculate level among obtained clusters. If fails reach pre-defined threshold, it is necessary adjustment modify DMs' preferences. At last, carried out obtain collective ranking alternatives. An illustrative example detailed simulation experiments demonstrated show validity against traditional models which just consider at only one stage for clustering.