Delegation Mechanism-Based Large-Scale Group Decision Making With Heterogeneous Experts and Overlapping Communities
针对大规模群体决策中专家群体存在重叠社区和内外异质专家的问题，提出基于社区检测和委托机制的方法，通过双参考点共识度量达成一致，并以5G投资案例验证。
The widespread utilization of novel technological and societal paradigms gave birth to the development large-scale group decision making. This study focuses on two issues regarding the structure of the large group. The first is about the overlapping clusters/communities (interdependent subgroups) of large-scale experts identified from the dimension reduction process. The second is regarding the situation in which internal and external experts (i.e., heterogeneous experts) exist in a large-scale group decision-making problem simultaneously. To manage the first issue, we propose to use an existing overlapping community detection method to cluster experts considering the independencies of communities. To address the second issue, a delegation mechanism is introduced to allocate the trust weights of external experts to internal experts. Furthermore, to reach a consensus for large-scale experts, a consensus measure based on two reference points (global collective preference and community preference) is given. Finally, an illustrative example regarding 5G technology investment is provided to verify the applicability of the proposed model.