摘要
在更高的抽象层次上总结了军事事件中多目标协同配合关系,使得基于事件分析的多目标协同配合模式成为可能。以“事件目标”关系为切入点,首先阐述了不同数据源的“事件目标”关系提取方法;其次,基于二分网络对事件和目标关系网建模,并以此生成目标关系加权单部投影图,采用Louvain模块度社团发现算法挖掘具有紧密联系的目标关系子图结构,形成多目标协同配合模式。仿真结果表明,利用二分网络建模“事件目标”关系,能有效解决非合作条件下数量未知的目标间协同配合关系分析的问题。基于关系权重的社团发现算法,可以在不同层次上发现多目标的协同配合组成模式,对多目标协同事件的预测和态势决策有重要意义。
This paper summarizes the multi-target coordination relationship in military event at a higher abstraction level,which makes it possible to analyze the multi-target coordination mode based on events.Taking the“event-target”relationship as the starting point,this paper firstly expounds the extraction methods of“event-target”relationship from different data sources;secondly,performs modeling to the event and target relationship network based on the bipartite network,and the weighted single projection drawing of target relationship is generated based on this.This paper uses the Louvain modularity community discovery algorithm to mine the subgraph structure of target relationships with closely connection,and forms multi-target cooperation mode.The simulation results show that using the bipartite network to model the“event-target”relationship can effectively solve the problem of analyzing the cooperative relationship among targets with unknown quantity under non-cooperative conditions.The community discovery algorithm based on relational weights can discover multi-target cooperation composite patterns at different levels,which is of great significance to the prediction and situational decision-making of multi-target cooperative events.
作者
何欢
李高云
旷生玉
江果
张伟
HE Huan;LI Gao-yun;KUANG Sheng-yu;JIANG Guo;ZHANG Wei(The 29th Research Institute of CETC,Chengdu 610036,China)
出处
《舰船电子对抗》
2022年第3期12-16,共5页
Shipboard Electronic Countermeasure
基金
联合基金项目,项目编号:U20B2070。
关键词
二分网络
社团发现
协同
军事事件
bipartite network
community discovery
cooperation
military event
