Modeling Multi-mapping Relations for Precise Cross-lingual Entity Alignment

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 813–822, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 813 Modeling Multi-mapping Relations for Precise Cross-lingual Entity Alignment Xiaofei Shi1, Yanghua Xiao1,2,3 1School of Computer Science, Fudan University, China 2CETC Big Data Research Institute Co., Ltd., Guizhou, China 3Big Data Application on lmproving Government Governance Capabilities National Engineering Laboratory, Guizhou, China {shixf18, shawyh}@fudan.edu.cn Abstract Entity alignment aims to find entities in differ- ent knowledge graphs (KGs) that refer to the same real-world object. An effective solution for cross-lingual entity alignment is crucial for many cross-lingual AI and NLP applications. Recently many embedding-based approaches were proposed for cross-lingual entity align- ment. However, almost all of them are based on TransE or its variants, which have been demonstrated by many studies to be unsuitable for encoding multi-mapping relations such as 1-N, N-1 and N-N relations, thus these meth- ods obtain low alignment precision. To solve this issue, we propose a new embedding-based framework. Through defining dot product- based functions over embeddings, our model can better capture the semantics of both 1-1 and multi-mapping relations. We calibrate em- beddings of different KGs via a small set of pre-aligned seeds. We also propose a weighted negative sampling strategy to generate valu- able negative samples during training and we regard prediction as a bidirectional problem in the end. Experimental results (especially with the metric Hits@1) on real-world multilingual datasets show that our approach significantly outperforms many other embedding-based ap- proaches with state-of-the-art performance. 1 Introduction Knowledge bases in the form of knowledge graphs (KGs) contain lots of facts of the real world. In recent years, KGs have been successfully ex- plored to serve many AI and NLP applications such as question answering, information extrac- tion, semantic search and recommender systems. Multilingual KGs such as BabelNet (Navigli and Ponzetto, 2012), YAGO3 (Mahdisoltani et al., 2013) and DBpedia (Lehmann et al., 2015) play essential roles in many cross-lingual applications. However, for multilingual KGs, each language- specific part is constructed by different parties with different data sources. Thus these language- specific KGs contain different but complementary facts. To support cross-lingual applications, it is essential for us to integrate these language-specific KGs into a unified KG. Fortunately, existing inter- lingual links (ILLs) that link equivalent entities or relations can help us to achieve this goal. How- ever, as Chen et al. (2017) and Sun et al. (2017) pointed out, the existing cross-lingual alignments usually account for a small proportion of the total. Therefore, it is valuable to find more cross-lingual entity alignment pairs to bridge the language gap in multilingual KGs. Entity alignment is the task of finding entity pairs in different KGs that refer to the same real- world object. An effective solution for entity alignment is vital for integrating multiple KGs. In this paper, we focus on cross-lingual entity align- ment. Our goal is to find more cross-lingual entity alignments based …

Content truncated to 500 words