基于信任关系和信息测度的概率语义社会网络群决策模型

设计专家权重和属性指标权重的计算模型已成为近年来备受关注的两个重要研究课题。针对评价信息为概率语义信任函数的社会网络群决策问题，提出一种基于信任关系和信息测度的概率语义社会网络群决策模型。首先，构建基于信任关系的概率语义决策空间，探究专家之间的信任传递模型，通过专家之间信任关系计算专家的权重；其次，引入概率语义信任函数的熵和相似度概念，并运用三角函数设计概率语义信任函数信息熵和相似度的衡量方法；最后，构建基于信任关系和信息测度的概率语义社会网络群决策模型，进而得到合理可靠的决策结果，同时将提出的社会网络群决策模型用于电动汽车供应商的选择实例，对比分析实验验证了模型的合理性和有效性。

Social Network Group Decision-Making Model Based on Trust Relationship and Information Measures with Probabilistic Linguistic Information

The probabilistic linguistic term sets are useful tool to address situations, in which the decision makers (DMs) are more comfortable providing their evaluation information linguistically rather than numerical values, and it is more reasonable and convenient to present the linguistic values by including an occurrence probability. Designing the calculation model of expert weights and attribute weights are two significant and challenging research issues in recent years. A probabilistic linguistic social network group decision-making model is investigated based on trust relationship and information measures to solve the group decision-making problem, in which evaluation information is a probabilistic linguistic trust function. Firstly, a probabilistic linguistic decision space based on trust relationship is constructed, and the trust propagation model among experts is explored to identify the incomplete trust relationship in the probabilistic linguistic trust relationship matrix. The weights of experts are calculated by using the trust relationship among experts. Then, two axiomatic definitions of information measures for probabilistic linguistic trust functions are presented, including the probabilistic linguistic trust function entropy and the probabilistic linguistic trust function similarity measure, to measure the uncertainty degree of probabilistic linguistic trust functions and the similarity degree between probabilistic linguistic trust functions. Subsequently, the calculating methods of probabilistic linguistic trust function’s entropy and similarity measure are designed by using triangular function. Finally, based on trust relationship and information measures, a probabilistic linguistic social network group decision-making model is constructed to derive the reasonable and reliable decision-making results. The proposed social network group decision-making model is applied to the selection of electric vehicle suppliers, and the comparison with existing approach is performed to validate the rationality and effectiveness of the proposed model.