基于动态图注意力聚合多跳邻域的实体对齐  

Entity alignment based on dynamic graph attention aggregation in multi-hop neighborhood

作　　者：汪浣沙 黄瑞阳[1,2] 宋旭晖 余诗媛 胡楠 Wang Huansha;Huang Ruiyang;Song Xuhui;Yu Shiyuan;Hu Nan(National Digital Switching System Engineering&Technological R&D Center,Zhengzhou 450002,China;Information Engineering University,Zhengzhou 450002,China;Software College,Zhengzhou University,Zhengzhou 450001,China)

机构地区：[1]国家数字交换系统工程技术研究中心,河南郑州450002 [2]中国人民解放军战略支援部队信息工程大学,河南郑州450002 [3]郑州大学软件学院,河南郑州450001

出　　处：《电子技术应用》2022年第11期51-56,共6页Application of Electronic Technique

基　　金：国家自然科学基金项目(62002384);中国博士后科学基金面上项目(2020M683760)。

摘　　要：实体对齐是实现对不同来源知识库进行融合的重要技术方法,在知识图谱、知识补全领域具有广泛应用。现有基于图注意力的实体对齐模型多使用静态图注意力网络且忽略了实体属性中的语义信息,导致模型存在有限注意、难以拟合、表达能力不足等问题。针对这些问题,开展基于动态图注意力结构建模实体对齐方法研究,首先使用图卷积层建模目标实体的单跳节点表示,其次应用动态图注意力网络获得多跳节点注意力系数并建模,再次利用逐层门控网络聚合图卷积层与动态图注意力层输出的单跳、多跳节点信息,最后拼接通过外部知识预训练自然语言模型提取的实体名称属性嵌入并进行相似度计算。该方法在DBP15K的三类跨语言数据集中都获得了一定的提高,证明了应用动态图注意力网络与融入实体属性语义在提高实体表示能力上的有效性。Entity alignment is an important technical method to realize the fusion of knowledge bases from different sources.It is widely used in the fields of knowledge graph and knowledge completion.The existing entity alignment models based on graph attention mostly use static graph attention network and ignore the semantic information in entity attributes,resulting in the problems of limited attention,difficult fitting and insufficient expression ability of the model.To solve these problems,this paper studies the entity alignment method based on the structure modeling of dynamic graph attention.Firstly,the single hop node representation of the target entity is modeled by GCN.Secondly,the multi hop node attention coefficient is obtained and entity modeled by using the dynamic graph attention network,and then the single hop and multi hop node information output by GCN and dynamic graph attention layer is aggregated by layer-wise gating network.Finally,the entity attribute semantic extracted by external knowledge pre training natural language model is embedded and concatenated to calculate similarity.This method has been improved in three types of cross language datasets of DBP15K,which proves the effectiveness of applying dynamic graph attention network and integrating entity attribute semantics in improving entity representation ability.

关 键 词：动态图注意力 图卷积网络 实体对齐 知识图谱 表示学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]