社会网络行为数据驱动的大群体应急决策共识模型

Consensus model for large group emergency decision-making driven by social network behavior data

针对大群体应急决策中决策参考信息不足和共识达成困难的问题,构建了一种由社会网络行为数据驱动的大群体应急决策共识模型。首先,针对决策参考信息不足的问题,提出基于公众行为大数据分析的属性挖掘方法。该方法将公众关注度作为数据聚类标准,挖掘社交媒体数据中的公众关注主题和权重,将其作为决策参考信息,应用到大群体应急决策过程。其次,针对大群体共识达成困难的问题,建立了基于信任-相似分析的决策共识模型。该模型在专家社会网络聚类的基础上,引入自信系数和保留系数,设计共识反馈规则以调整专家偏好,提高了群体共识水平。最后,通过公共安全事件案例验证了所提模型的可行性和合理性。

Aiming at the problems of insufficient decision-making reference information and difficulty in reaching consensus in large-group emergency decision-making,a large-group emergency decision-making consensus model driven by social network behavior data is constructed.Firstly,in view of the lack of decision-making reference information,an attribute mining method based on public behavior big data analysis is proposed.The proposed method takes public attention as the data clustering standard,and mines the topics and weights of public concern topic in social media data as decision reference information to participate in the large group emergency decision-making process.Secondly,considering the difficulty in reaching consensus in large group,a decision consensus model based on trust-similarity analysis is established.Based on expert social network clustering,the proposed model introduces confidence coefficient and retention coefficient,and designs consensus feedback rules to adjust experts’preferences and improves the level of group consensus.As a result,the proposed model is applied to a public security incident case to verify its feasibility and rationality.