ChEMU-Ref: A Corpus for Modeling Anaphora Resolution in the Chemical Domain

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 1362–1375 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 1362 ChEMU-Ref: A Corpus for Modeling Anaphora Resolution in the Chemical Domain Biaoyan Fang1, Christian Druckenbrodt2, Saber A. Akhondi2, Jiayuan He1,3, Timothy Baldwin1 and Karin Verspoor1 1The University of Melbourne, Australia 2Elsevier 3RMIT University, Australia biaoyanf@student.unimelb.edu.au {c.druckenbrodt,s.akhondi}@elsevier.com {estrid.he,tbaldwin,karin.verspoor}@unimelb.edu.au Abstract Chemical patents contain rich coreference and bridging links, which are the target of this re- search. Specially, we introduce a novel anno- tation scheme, based on which we create the ChEMU-Ref dataset from reaction description snippets in English-language chemical patents. We propose a neural approach to anaphora resolution, which we show to achieve strong results, especially when jointly trained over coreference and bridging links. 1 Introduction Chemical research has contributed greatly to human society and wellbeing, including new medicines and vaccines (Gwynne and Heabrer, 2015). Research is heavily reliant on knowledge of existing chemical processes and methods of chemi- cal synthesis, which are documented in chemical research literature and chemical patents. Given the rapid growth of both publications and patents in chemistry, the need for automatic methods to extract semi-structured knowledge from chemical texts is becoming increasingly critical (Li et al., 2016; Akhondi et al., 2019). Anaphora resolution is a key component of com- prehensive information extraction (R¨osiger, 2019; Poesio et al., 2016). In chemistry, different chem- ical compounds are mixed and reacted together in different ways to generate novel compounds, and to understand the precise chemical process of- ten involves both resolving anaphoric references and understanding chemical changes/interactions a given entity is involved in. For example, as seen in Figure 1, while the final mention of mixture on line 3 and that on line 4 are both coreferent and chemi- cally identical, in the case of mixture on line 2 and the first mention of mixture on line 3, the chemical composition is the same but a transformation has taken place via the stir and cool actions. Our aim in this paper is to both identify anaphoric references in chemical patents, and de- termine the chemical relation between each linked pair of entities. We propose a domain-specific an- notation framework based on five types of anaphora relations combining coreference and bridging. We then construct a dataset following this framework, annotated by chemical experts who achieve high inter-annotator agreement. We additionally extend existing anaphora resolution methods to model anaphora in chemical text, and compare both component-wise and joint models for anaphora res- olution. This dataset will be released as part of the upcoming ChEMU 2021 shared task1 (He et al., to appear). Our contributions in this paper are as follows: (1) we propose a novel annotation scheme for anaphora resolution in chemical patents; (2) we develop a novel anaphora-resolution dataset based on chem- ical patents; and (3) we extend a general-purpose coreference resolution method, and achieve strong results via joint training over coreference and bridg- ing with domain-specific fine-tuning. 2 Related Work Anaphora occurs …

Content truncated to 500 words