基于军事新闻图谱的事理路径寻优方法  

An Event Paths Optimization Method Based on Knowledge Graph of Military News

在线阅读下载全文

作  者:房湧淇 张中杰 王冬 黄健[1] FANG Yong-qi;ZHANG Zhong-jie;WANG Dong;HUANG Jian(College of Intelligent Science,National University of Defense Technology,Changsha Hunan 410003,China)

机构地区:[1]国防科技大学智能科学学院,湖南长沙410003

出  处:《计算机仿真》2024年第4期1-7,共7页Computer Simulation

基  金:国家自然科学基金资助项目(61906202)。

摘  要:基于事理图谱的事理路径分析是辅助决策的重要手段,现有的重要事理路径寻优方法中对于事理路径价值评估指标的设计较为单一,大多只是以输出最短路径作为路径寻优的目标。为能够更加充分地反映事理路径的重要性信息,根据节点重要性、路径稀缺性和关系边得分情况三方面提出新型路径价值评估函数,以丰富事理路径价值评估方法。同时为解决经典路径搜索算法应用场景单一、搜索效率低的问题,基于Dijkstra算法设计偏离路径搜索算法,旨在满足路径寻优的不同场景,并提高搜索效率。最后采用自定义的国际军事新闻事理图谱作为数据集进行实验,结果显示上述搜索算法在运行内存占用方面能够达到与经典路径搜索算法同等的水平,并在搜索速度上有所提升,输出的重要性路径满足预设要求。The event path analysis based on the event graph is an important means to assist decision-making.In the existing important event path optimization methods,the design of the evaluation index of the event path value is relatively simple,and most of them only take the output of the shortest path as the goal of path optimization.In order to more fully reflect the importance information of the event path,this paper proposes a new path value evaluation function based on three aspects:node importance,path scarcity and relationship edge score to enrich the event path value evaluation method.At the same time,in order to solve the problem of single application scenario and low search efficiency of the classical path search algorithm,a deviated path algorithm was designed based on the Dijkstra algorithm,which aims to meet different scenarios of path optimization and improve the search efficiency.Finally,a custom map of international military news affairs was used as a data set to conduct experiments.The results show that the search algorithm can reach the same level as the classic path search algorithm in terms of running memory usage,and the search speed has been improved.The importance of output The path meets the preset requirements.

关 键 词:事理图 路径规划 价值评估函数 路径搜索算法 路径重要性 

分 类 号:TP391.9[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象