Knowledge Graph Alignment with Entity-Pair Embedding

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 1672–1680, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 1672 Knowledge Graph Alignment with Entity-Pair Embedding Zhichun Wang, Jinjian Yang, Xiaoju Ye School of Artificial Intelligence, Beijing Normal University, Beijing, China Engineering Research Center of Intelligent Technology and Educational Application, Ministry of Education, Beijing, China zcwang@bnu.edu.cn {tecyyang,mryx}@mail.bnu.edu.cn Abstract Knowledge Graph (KG) alignment is to match entities in different KGs, which is important to knowledge fusion and integration. Recently, a number of embedding-based approaches for KG alignment have been proposed and achieved promising results. These approaches first embed entities in low-dimensional vec- tor spaces, and then obtain entity alignments by computations on their vector representa- tions. Although continuous improvements have been achieved by recent work, the per- formances of existing approaches are still not satisfactory. In this work, we present a new approach that directly learns embeddings of entity-pairs for KG alignment. Our approach first generates a pair-wise connectivity graph (PCG) of two KGs, whose nodes are entity- pairs and edges correspond to relation-pairs; it then learns node (entity-pair) embeddings of the PCG, which are used to predict equivalent relations of entities. To get desirable embed- dings, a convolutional neural network is used to generate similarity features of entity-pairs from their attributes; and a graph neural net- work is employed to propagate the similarity features and get the final embeddings of entity- pairs. Experiments on five real-world datasets show that our approach can achieve the state- of-the-art KG alignment results. 1 Introduction Knowledge graphs (KGs) have been built and ap- plied in several domains, including question an- swering (Zhang et al., 2018), recommendation (Sun et al., 2018b), and information extraction (Yang and Mitchell, 2017). Most existing KGs are built sep- arately by different organizations, using different data sources and languages. Therefore, KGs are heterogeneous that the same entity may exist in dif- ferent KGs in different surface forms. On the other hand, KGs can be complementary to each other; knowledge about the same entity may distribute in several KGs. To handle the heterogeneity prob- lem and integrate knowledge in different KGs, it is essential to perform KG alignment, i.e. matching entities in separate KGs. Recently, KG embedding models have been ex- plored in solving the problem of KG alignment. A number of embedding-based approaches have been proposed, including MTransE (Chen et al., 2017), JAPE (Sun et al., 2017), IPTransE (Zhu et al., 2017), GCN-Align (Wang et al., 2018), RDGCN (Wu et al., 2019), and MultiKE (Zhang et al., 2019), etc. These approaches first em- bed entities in low-dimensional vector spaces, and then obtain the entity alignments by computa- tions on their vector representations. Compar- ing with traditional similarity-based approaches, embedding-based ones can effectively model dif- ferent kinds of information in KGs, which align entities without manually designed similarity fea- tures. Most recently, continuous improvements have been achieved by combining multiple kinds of information in KGs or using more sophisticated embedding models. However, the performances of most approaches are …

Content truncated to 500 words