arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.10873cs.CLcs.FL
收藏

知识图谱实体分类的连接嵌入

Connecting Embeddings for Knowledge Graph Entity Typing

Southwestern University of Finance and Economics(西南财经大学);Carnegie Mellon University(卡内基梅隆大学);Tencent(腾讯);University of Chinese Academy of Sciences(中国科学院大学)

Yu Zhao, Anxiang Zhang, Ruobing Xie, Kang Liu, Xiaojie Wang

摘要:

知识图谱实体分类的目的是推断知识图谱中可能缺失的实体类型实例,这是知识图谱完成的一个非常重要但仍有待开发的子任务。在本文中,我们提出了一种新的KG实体类型的训练方法,通过联合利用本地类型知识从现有的实体类型断言和全球三元知识从KG。具体来说,我们提出了两个不同的知识驱动的实体类型推断的有效机制。因此,我们建立了两个新的嵌入模型来实现的机制。然后,与他们的联合模型被用来推断丢失的实体类型实例,这有利于推理,同意实体类型实例和知识库中的三元知识。在两个真实数据集(Freebase和YAGO)上的实验结果表明,我们提出的机制和模型可以有效地提高KG实体类型。本文的源代码和数据可以从:https://github.com/ Adam 1679/ConnectE

英文摘要:

Knowledge graph (KG) entity typing aims at inferring possible missing entity type instances in KG, which is a very significant but still under-explored subtask of knowledge graph completion. In this paper, we propose a novel approach for KG entity typing which is trained by jointly utilizing local typing knowledge from existing entity type assertions and global triple knowledge from KGs. Specifically, we present two distinct knowledge-driven effective mechanisms of entity type inference. Accordingly, we build two novel embedding models to realize the mechanisms. Afterward, a joint model with them is used to infer missing entity type instances, which favors inferences that agree with both entity type instances and triple knowledge in KGs. Experimental results on two real-world datasets (Freebase and YAGO) demonstrate the effectiveness of our proposed mechanisms and models for improving KG entity typing. The source code and data of this paper can be obtained from: https://github.com/ Adam1679/ConnectE

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑