基于属性序的贝叶斯网结构快速学习方法  

Rapid method to build Bayesian network structure based on attribute order relation

作　　者：吕宗磊 芦穆庆[2] LYU Zong-lei;LU Mu-qing(Information Technology Research Base of Civil Aviation Administration of China,Tianjin 300300,China;College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航信息技术科研基地,天津300300 [2]中国民航大学计算机科学与技术学院,天津300300

出　　处：《计算机工程与设计》2018年第9期2961-2966,共6页Computer Engineering and Design

基　　金：国家科技支撑计划课题基金项目(2014BAJ04B02)

摘　　要：提出一种基于属性序的贝叶斯网结构快速学习方法,实现贝叶斯网结构的快速构建。以先验知识和样本数据信息为基础进行贝叶斯网结构模型的构建;在此基础上进行参数学习,得到体现各个节点相互依赖关系的贝叶斯网。实验结果表明,该方法可以有效学习贝叶斯网结构,训练得到的贝叶斯网络节点的后验概率与统计得到的条件概率相当接近,可以用于有效分析航班保障各作业环节之间的相关性。A rapid method to build Bayesian network structure based on attribute order relation was presented.The construction of Bayesian network was achieved rapidly.The Bayesian network structure model based on prior knowledge and sample data information was built.The parameters were learned based on the network structure.The Bayesian network that characterizing the relationship between the nodes was obtained.Experimental results show that the proposed method can effectively study the Bayesian network structure.The posteriori probability of the Bayesian network nodes trained using the method is quite close to the conditional probability.The method proposed can be used to analyze the correlation between the flight support links.

关 键 词：贝叶斯网 结构学习 属性序 机器学习 航班保障 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]