基于云模型与证据理论的共识分析方法

Approach for analyzing consensus based on cloud model and evidence theory

针对综合集成研讨厅中发散型研讨具有连续性、模糊性、复杂性等特点,提出了一种基于云模型与证据理论的共识分析方法.该方法利用云模型根据研讨信息间语义关系求取每条发言表达的意见;并利用证据理论根据某专家所有时刻发言表达的意见推理该专家意见;最后根据求得的各专家意见形成定性的专家意见分布图和定量的共识水平、关注水平.实例表明该方法是有效的,它能使研讨专家实时了解当前共识状态,提高群体研讨效率.

An approach for analyzing consensus based on cloud model and evidence theory is proposed, with respect to the characteristics of divergent argument in HWME, such as continuity, fuzziness and complexity. Firstly, this method uses cloud model to find out the opinion implicated in every comment, according to the semantic relations between argument information; then synthesizes the implicated opinions expressed by an expert in all time into the expert opinion on the base of evidence theory; and at last, obtains qualitative distribution map of expert opinion and quantitative level of consensus and attention, in accordance with acquired opinions from different experts. A numerical example indicates that this approach is effective, enables the experts to understand current consensus state and improves efficiency of the group argument.