%0 Journal Article
%T 基于循环一致性损失的知识图谱嵌入的研究<br>A Study of Knowledge Graph Embedding Based on Cyclic Consistency
%A 李迦龙
%A 郭中华
%A 蔺金元
%J Artificial Intelligence and Robotics Research
%P 822-836
%@ 2326-3423
%D 2024
%I Hans Publishing
%R 10.12677/airr.2024.134085
%X 基于生成对抗网络中的循环一致性原则&#65292;提出了一个基于循环一致性损失的知识图谱嵌入模型。该模型首先使用ConvE模型利用头实体和关系构造的&#8220;图片&#8221;对尾实体进行预测&#65292;再利用尾实体和关系构造的&#8220;逆图片&#8221;对头实体进行预测。同时根据循环一致性原理&#65292;构造了ConvE模型的一个新的损失函数&#65292;解决了网络的可逆性。在WN18、FB15k以及YAGO3-10三个数据集上设计实验&#65292;证明了模型有效地缩短了头实体和原头实体的语义空间距离。<br>Based on the cyclic consistency principle in generative adversarial networks, a knowledge graph embedding model based on cyclic consistency loss is proposed. Firstly, the ConvE model is used to predict the tail entity by using the &#8220;picture&#8221; constructed by the head entity and the relationship, and then the &#8220;inverse picture&#8221; constructed by the tail entity and the relationship is used to predict the head entity. According to the principle of cyclic consistency, a new loss function of ConvE model is constructed to solve the reversibility of the network. Experiments are designed on WN18, FB15k and YAGO3-10 data sets, and it is proved that the model can effectively shorten the semantic space distance between the header entity and the original header entity.
%K 知识图谱&#65292
%K TransE&#65292
%K ConvE&#65292
%K 循环一致性<br>Knowledge Graph
%K TransE
%K ConvE
%K Cyclic Consistency
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=101145