基于贝叶斯网络的民航突发事件因果关系分析方法研究

Research on causal reasoning method of domain ontology based on Bayesian network

针对民航突发事件因果关系无法有效评估与关联分析的问题,提出了一种基于贝叶斯网络的民航突发事件因果关系分析方法。在民航突发事件应急管理领域本体的基础上引入贝叶斯理论,通过规则设计实现了领域本体中概念、关系与实例的贝叶斯网络转换;然后采用贝叶斯网络知识合成算法E-IPFP构建贝叶斯网络节点的条件概率表,并通过消息传递机制计算父子节点间的概率关系,获得民航突发事件因果关系的概率分布。采用民航突发事件应急管理领域本体和世界民航事故调查跟踪报告中的案例作为实验数据,给出了民航突发事件因果间关系的分析,为基于大数据的突发事件关联分析与推理提供了方法支持。

According to the fact that civil aviation emergency cannot effectively evaluate causality and correlation analysis,this paper proposed a causal analysis method for civil aviation emergencies based on Bayesian network.The method introduced Bayesian theory on the basis of the domain ontology of civil aviation emergency management.Firstly,the Bayesian network realized the transformation of concepts,relations and instances in domain ontology with rules design.Then it constructed the conditional probability table for Bayesian network nodes using Bayesian network knowledge synthesis algorithm E-IPFP,and calculated the probability relationship between parent nodes and child nodes through the message passing mechanism,to obtain the probability distribution of civil aviation incidents causation.This paper adopted the civil aviation emergency management domain ontology and the world civil aviation accident investigation tracking report as experimental data,gave the analysis of the causal relationship between civil aviation emergencies,which provided the method to support the correlation analysis and reasoning of unexpected events based on large data.