Abstract: The purpose of this paper is to solve the problem that the trust relationship between the large group emergency decision-making experts and the decision-making risk are caused by the transmission, as well as the problem of generating independent aggregation resulting from the large differences in individual preferences among large groups. Firstly, a “trust-knowledge model” is proposed to integrate and transfer the trust relationship between decision-making experts, and the trust knowledge network between decision-making experts is obtained according to the trust risk preference of decision-making experts. Secondly, the network of trust knowledge degree is clustered by Louvain algorithm to obtain several aggregations efficiently and quickly, and we use the technology of social network analysis to determine the weight of each decision maker and aggregation. Then we aggregate the decision makers' preferences in each aggregation, sorting the information given by the decision makers to rank the alternative decision schemes. Finally, the rationality and effectiveness of the proposed method are verified by a case analysis and a comparative study.

Key words: large group, risk decision, trust network, social network, trust delivery

摘要： 针对大群体应急决策专家之间信任关系及其传递引发的决策风险,以及由于大群体中个体偏好差异较大导致生成独立聚集等问题。首先,提出一个“信任—知识模型”对决策专家之间的信任关系进行集成和传递,并根据决策专家的信任风险偏好得出决策专家之间的信任知识度网络;其次,利用Louvain算法对信任知识度网络进行聚类,高效快速的获得若干个聚集,并用社会网络分析技术确定每个决策者和聚集的权重;然后对每个聚集中的决策者偏好进行集结,并综合决策者给出的信息对备选决策方案进行排序。最后,通过案例分析和对比验证了所提方法的合理性与有效性。

关键词: 大群体, 风险决策, 信任网络, 社会网络, 信任传递