摘要
通过全面分析当前装备知识图谱补全的问题和国内外研究现状,提出一种基于注意力机制的装备知识图谱内部链接关系补全方法。在仿真过程中,针对传统方法中属性特征缺失和语义信息不足的问题,通过图注意力模型学习实例节点周围丰富的语义信息,并利用卷积网络实现图谱中各类嵌入表示的全局交互,最终实现装备链接关系的补全。
By comprehensively analyzing the current problems of equipment knowledge graph completion and the research status at home and abroad,an internal link relationship completion method of equipment knowledge graph based on attention mechanism is proposed.To address missing attribute features and insufficient semantic information in traditional methods during the simulation process,the rich semantic information surrounding the instance nodes is learned by the graph attention model,then the convolutional network is utilized to realize the global interaction of various types of embedded representations in the graph,and the completion of the equipment link relationships is finally realized.
作者
范鹏
杨启民
李文桥
FAN Peng;YANG Qimin;LI Wenqiao(Unit 93209 of PLA,Beijing 100085,China;Computer School,Beijing Information Science and Technology University,Beijing 100101,China)
出处
《火力与指挥控制》
北大核心
2025年第9期105-111,共7页
Fire Control & Command Control
关键词
装备知识图谱
链接关系补全
深度学习
注意力机制
equipment knowledge graph
link relationship completion
deep learning
attention mechanism
