Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 6355–6364, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 6355 Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment Zhiyuan Liu12 Yixin Cao2∗ Liangming Pan12 Juanzi Li3 Zhiyuan Liu3 Tat-Seng Chua2 1NUS Graduate School for Integrative Sciences and Engineering 2School of Computing, National University of Singapore, Singapore 3Department of CST, Tsinghua University, Beijing, China {acharkq, caoyixin2011}@gmail.com, e0272310@u.nus.edu {lijuanzi, liuzy}@tsinghua.edu.cn, dcscts@nus.edu.sg Abstract Entity alignment (EA) aims at building a uni- fied Knowledge Graph (KG) of rich content by linking the equivalent entities from various KGs. GNN-based EA methods present promis- ing performance by modeling the KG struc- ture defined by relation triples. However, at- tribute triples can also provide crucial align- ment signal but have not been well explored yet. In this paper, we propose to utilize an attributed value encoder and partition the KG into subgraphs to model the various types of attribute triples efficiently. Besides, the per- formances of current EA methods are overes- timated because of the name-bias of existing EA datasets. To make an objective evaluation, we propose a hard experimental setting where we select equivalent entity pairs with very dif- ferent names as the test set. Under both the regular and hard settings, our method achieves significant improvements (5.10% on average Hits@1 in DBP15k) over 12 baselines in cross- lingual and monolingual datasets. Ablation studies on different subgraphs and a case study about attribute types further demonstrate the effectiveness of our method. Source code and data can be found at https://github.com/ thunlp/explore-and-evaluate. 1 Introduction The prosperity of data mining has spawned Knowl- edge Graphs (KGs) in many domains that are often complementary to each other. Entity Alignment (EA) provides an effective way to integrate the com- plementary knowledge in these KGs into a unified KG by linking equivalent entities, thus benefiting knowledge-driven applications such as Question Answering (Yang et al., 2017, 2018), Recommen- dation (Cao et al., 2019b) and Information Extrac- tion (Kumar, 2017; Cao et al., 2018). However, EA is a non-trivial task that it could be formulated as *Corresponding author. a quadratic assignment problem (Yan et al., 2016), which is NP-complete (Garey and Johnson, 1990). A KG comprises a set of triples, with each triple consisting of a subject, predicate, and ob- ject. There are two types of triples: (1) relation triples, in which both the subject and object are en- tities, and the predicate is often called relation (see Figure 1(a)); and (2) attribute triples, in which the subject is an entity and the object is a value, which is either a number or literal string (see Figure 1(c)), and the predicate is often called attribute. Most of the previous EA models (Sun et al., 2017; Wang et al., 2018; Wu et al., 2019a) rely on the structure assumption that, the adjacencies of two equivalent entities in KGs usually contain equivalent entities (Wang et al., 2018) (see Fig- ure 1(a)). These models mainly focus on modeling KG structure defined …

Content truncated to 500 words