基于贝叶斯网络的突发事件信息生命阶段研判方法

Judged Method of Emergency Information Life Stages Based on Bayesian Networks

突发事件发生后,明确信息的实时生命阶段对于建立有效且操作性强的应急信息管理措施至关重要。突发事件信息在传播过程中表现出生命周期特性,将突发事件信息生命过程划分为初生萌芽期、快速成长期、衰退死亡期三个阶段。将影响信息传播的关键要素抽象到贝叶斯网络模型中并给出贝叶斯网络结构学习方法,构建了突发事件信息生命阶段研判模型。最后,结合历史突发事件的有关数据进行Netica仿真,得出的结果能够实时、动态的判断突发事件信息所处的生命阶段,为政府进行突发事件应急信息管理提供理论参考。

After the occurrence of emergency invents, it＇s significant for government to realize the life stage of emergencyinformation. According to information life cycle theory,the emergency incidents＇ life cycly can be divided into three stages：birth period,growth period and death period. The critical elements of information dissemination are extracted into aBayesian network judge model quantitatively; the learning schemes of Bayesian network are also given then. Finally, Neticasimulation with recent years＇ emergency events＇ data is accomplished. The results prove that this model can provide anaccurate judgment of emergency events＇ life stage at any time and dynamically. This can help government to prepare forcorresponding management immediately whenever emergency event happens.