基于循环一致性损失的知识图谱嵌入的研究
A Study of Knowledge Graph Embedding Based on Cyclic Consistency
DOI: 10.12677/airr.2024.134085, PDF,    科研立项经费支持
作者: 李迦龙, 郭中华*, 蔺金元：宁夏大学物理与电子电气工程学院，宁夏 银川
关键词: 知识图谱；TransE；ConvE；循环一致性；Knowledge Graph； TransE； ConvE； Cyclic Consistency
摘要: 基于生成对抗网络中的循环一致性原则，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“图片”对尾实体进行预测，再利用尾实体和关系构造的“逆图片”对头实体进行预测。同时根据循环一致性原理，构造了ConvE模型的一个新的损失函数，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，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。
Abstract: 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 “picture” constructed by the head entity and the relationship, and then the “inverse picture” 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.
文章引用：李迦龙, 郭中华, 蔺金元. 基于循环一致性损失的知识图谱嵌入的研究[J]. 人工智能与机器人研究, 2024, 13(4): 822-836. https://doi.org/10.12677/airr.2024.134085

参考文献

[1] 张正航, 钱育蓉, 行艳妮, 等. 基于TransE的表示学习方法研究综述[J]. 计算机应用研究, 2021(3): 656-663.
[2] 昌攀, 曹扬. 改进的TransH模型在知识表示与推理领域的研究[J]. 广西大学学报(自然科学版), 2020, 45(2): 321-327.
[3] Zhang, Z., Jia, J., Wan, Y., Zhou, Y., Kong, Y., Qian, Y., et al. (2021) TransR*: Representation Learning Model by Flexible Translation and Relation Matrix Projection. Journal of Intelligent & Fuzzy Systems, 40, 10251-10259. [Google Scholar] [CrossRef] 
[4] Scott, S.D. and Choi, K.H. (2022) MicroRNA-122 and Poly-C Binding Protein-2 Regulate Hepatitis C Replication by Binding to Overlapping Sites on the 5’ Untranslated Region of the Viral Genome. Biophysical Journal, 121, 206a-207a. [Google Scholar] [CrossRef] 
[5] 余晓鹏, 何儒汉, 黄晋, 等. 基于改进Inception结构的知识图谱嵌入模型[J]. 计算机应用, 2022, 42(4): 1065-1071.
[6] Pornprasit, C., Kiattipadungkul, P., Duangkaew, P., Tuarob, S. and Noraset, T. (2020) Enhancing CNN Based Knowledge Graph Embedding Algorithms Using Auxiliary Vectors: A Case Study of Wordnet Knowledge Graph. 2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), Phuket, 24-27 June 2020, 763-766. [Google Scholar] [CrossRef] 
[7] Pang, L. (2022) Intelligent Big Information Retrieval of Smart Library Based on Graph Neural Network (GNN) Algorithm. Computational Intelligence and Neuroscience, 2022, Article ID: 1475069. [Google Scholar] [CrossRef] [PubMed]
[8] Gao, J., Liu, X., Chen, Y. and Xiong, F. (2022) MHGCN: Multiview Highway Graph Convolutional Network for Cross-Lingual Entity Alignment. Tsinghua Science and Technology, 27, 719-728. [Google Scholar] [CrossRef] 
[9] Zheng, Z., Li, J., Zhu, L., Li, H., Petzold, F. and Tan, P. (2022) GAT-CADNet: Graph Attention Network for Panoptic Symbol Spotting in CAD Drawings. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, 18-24 June 2022, 11737-11746. [Google Scholar] [CrossRef] 
[10] 饶国政, 许国顺. 基于实体与关系聚合图的知识图谱嵌入模型[P]. 中国专利, CN114564623A. 2022-03-10.
[11] 袁立宁, 李欣, 王晓冬, 等. 图嵌入模型综述[J]. 计算机科学与探索, 2022, 16(1): 59-87.

为你推荐