ELDEN: Improved Entity Linking Using Densified Knowledge Graphs

Proceedings of NAACL-HLT 2018, pages 1844–1853 New Orleans, Louisiana, June 1 - 6, 2018. c⃝2018 Association for Computational Linguistics ELDEN: Improved Entity Linking using Densified Knowledge Graphs Priya Radhakrishnan+, Partha Talukdar* and Vasudeva Varma+ +IIIT Hyderabad, India *Indian Institute of Science, Bangalore, India priya.r@research.iiit.ac.in, ppt@iisc.ac.in, vv@iiit.ac.in Abstract Entity Linking (EL) systems aim to automati- cally map mentions of an entity in text to the corresponding entity in a Knowledge Graph (KG). Degree of connectivity of an entity in the KG directly affects an EL system’s abil- ity to correctly link mentions in text to the entity in KG. This causes many EL systems to perform well for entities well connected to other entities in KG, bringing into focus the role of KG density in EL. In this paper, we propose Entity Linking using Densified Knowledge Graphs (ELDEN). ELDEN is an EL system which first densifies the KG with co-occurrence statistics from a large text cor- pus, and then uses the densified KG to train entity embeddings. Entity similarity measured using these trained entity embeddings result in improved EL. ELDEN outperforms state- of-the-art EL system on benchmark datasets. Due to such densification, ELDEN performs well for sparsely connected entities in the KG too. ELDEN’s approach is simple, yet effec- tive. We have made ELDEN’s code and data publicly available. 1 Introduction Entity Linking (EL) is the task of mapping men- tions of an entity in text to the corresponding entity in Knowledge Graph (KG) (Hoffart et al., 2011; Dong et al., 2014; Chisholm and Hachey, 2015). EL systems primarily exploit two types of infor- mation: (1) similarity of the mention to the candi- date entity string, and (2) coherence between the candidate entity and other entities mentioned in the vicinity of the mention in text. Coherence es- sentially measures how well the candidate entity is connected, either directly or indirectly, with other KG entities mentioned in the vicinity (Milne and Witten, 2008; Globerson et al., 2016). In the state- ✓ KNOWLEDGE GRAPH Andrei Broder won best paper award at WWW TEXT WEB CORPUS World_Wide_Web WWW_conference Program Committee PERTH, Australia, 5 April 2017: ... affiliation are Andrei Broder (Google/AltaVista),.. WWW9 Program Committee... WWW conference with more than 3,500 citations…. ? ? Andrei_Broder Figure 1: Improving entity disambiguation per- formance using KG densification: edges to WWW conference, a sparsely-connected entity in the KG, is increased by adding edges from pseudo entity Program Committee whose mention co- occurs with it in web corpus. ELDEN, the system proposed in this paper, uses such densified KG to successfully link ambiguous mention WWW to the correct entity WWW conference, instead of the more popular entity World Wide Web. of-the-art EL system by (Yamada et al., 2016), co- herence is measured as distance between embed- dings of entities. This system performs well on entities which are densely-connected in KG, but not so well on sparsely-connected entities in the KG. We demonstrate this problem using the exam- ple sentence in Figure 1. This sentence has two mentions: Andrei Broder …

Content truncated to 500 words