摘要
本文首先从战场态势与事件间的因果关系出发,构建了用于态势评估的贝叶斯网络推理模型;分析了联合树算法在态势评估中所存在的问题,结合基于D分离的相关推理在模型结构挖掘方面的特性,提出了一种联合树相关推理方法;最后,通过实验验证了该方法的有效性和准确性,为态势评估下的精确推理提供了一种有效的解决方案.
In this paper, first a Bayesian network reasoning model for situation assessment is built according to the causal link between the battlefield situation and events. Then a problem occurred when applying junction tree algorithm to situation assessment is analyzed. Combining with the characteristics of relevance reasoning based D-- Separation in model structure mining, a junction tree relevance reasoning method is proposed. Finally, the validity and accuracy of the method is verified by experiment, providing an effective solution of precise reasoning under situation assessment.
出处
《微电子学与计算机》
CSCD
北大核心
2013年第9期26-29,共4页
Microelectronics & Computer
基金
国家十二五探索项目(8134226)
关键词
态势评估
贝叶斯网络
联合树
相关推理
situation assessment~ Bayesian network~ junction tree
relevance reasoning
作者简介
张海生男，（1986-），硕士研究生．研究方向为战场态势评估、贝叶斯网络推理．
缪禅娜女，（1988-），硕士研究生．研究方向为战场态势评估下的时间序列预测．
