Representation Learning with Ordered Relation Paths for Knowledge Graph Completion

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 2662–2671, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 2662 Representation Learning with Ordered Relation Paths for Knowledge Graph Completion Yao Zhu1, Hongzhi Liu2∗, Zhonghai Wu3,4, Yang Song5 and Tao Zhang5 1 Center for Data Science, Peking University, Beijing, China 2 School of Software and Microelectronics, Peking University, Beijing, China 3 National Engineering Center of Software Engineering, Peking University, Beijing, China 4 Key Lab of High Confidence Software Technologies (MOE), Peking University, Beijing, China 5 BossZhipin NLP Center @Kanzhun.com, Beijing, China {yao.zhu,liuhz,wuzh}@pku.edu.cn, {songyang,kylen.zhang}@kanzhun.com Abstract Incompleteness is a common problem for ex- isting knowledge graphs (KGs), and the com- pletion of KG which aims to predict links be- tween entities is challenging. Most existing KG completion methods only consider the di- rect relation between nodes and ignore the re- lation paths which contain useful information for link prediction. Recently, a few methods take relation paths into consideration but pay less attention to the order of relations in paths which is important for reasoning. In addition, these path-based models always ignore non- linear contributions of path features for link prediction. To solve these problems, we pro- pose a novel KG completion method named OPTransE. Instead of embedding both entities of a relation into the same latent space as in previous methods, we project the head entity and the tail entity of each relation into differ- ent spaces to guarantee the order of relations in the path. Meanwhile, we adopt a pooling strategy to extract nonlinear and complex fea- tures of different paths to further improve the performance of link prediction. Experimental results on two benchmark datasets show that the proposed model OPTransE performs bet- ter than state-of-the-art methods. 1 Introduction Knowledge graphs (KGs) are built to store struc- tured facts which are encoded as triples, e.g., (Beijing, CapitalOf, China) (Lehmann et al., 2015). Each triple (h, r, t) consists of two entities h, t and a relation r, indicating there is a relation r between h and t. Large-scale KGs such as YAGO (Suchanek et al., 2007), Freebase (Bollacker et al., 2008) and WordNet (Miller, 1995) contain billions of triples and have been widely applied in vari- ous fields (Riedel et al., 2013; Dong et al., 2015). However, a common problem with these KGs is ∗Corresponding author that they are far from complete, which has limited the development of KG’s applications. Thus, KG completion with the goal of filling in missing parts of the KG has become an urgent issue. Specifi- cally, KG completion aims to predict whether a re- lationship between two entities is likely to be true, which is defined as the link prediction in KGs. Most existing KG completion methods are based on representation learning, which embed both entities and relations into continuous low- dimension spaces. TransE (Bordes et al., 2013) is one of the most classical KG completion models, which embeds entities …

Content truncated to 500 words