An Effective Transition-based Model for Discontinuous NER

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5860–5870 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 5860 An Effective Transition-based Model for Discontinuous NER Xiang Dai1,2 Sarvnaz Karimi1 Ben Hachey3 Cecile Paris1 1CSIRO Data61, Sydney, Australia 2University of Sydney, Sydney, Australia 3Harrison.ai, Sydney, Australia {dai.dai,sarvnaz.karimi,cecile.paris}@csiro.au ben.hachey@gmail.com Abstract Unlike widely used Named Entity Recognition (NER) data sets in generic domains, biomed- ical NER data sets often contain mentions consisting of discontinuous spans. Conven- tional sequence tagging techniques encode Markov assumptions that are efficient but pre- clude recovery of these mentions. We pro- pose a simple, effective transition-based model with generic neural encoding for discontinu- ous NER. Through extensive experiments on three biomedical data sets, we show that our model can effectively recognize discontinuous mentions without sacrificing the accuracy on continuous mentions. 1 Introduction Named Entity Recognition (NER) is a critical com- ponent of biomedical natural language processing applications. In pharmacovigilance, it can be used to identify adverse drug events in consumer reviews in online medication forums, alerting medication developers, regulators and clinicians (Leaman et al., 2010; Sarker et al., 2015; Karimi et al., 2015b). In clinical settings, NER can be used to extract and summarize key information from electronic med- ical records such as conditions hidden in unstruc- tured doctors’ notes (Feblowitz et al., 2011; Wang et al., 2018b). These applications require identi- fication of complex mentions not seen in generic domains (Dai, 2018). Widely used sequence tagging techniques (flat model) encode two assumptions that do not always hold: (1) mentions do not nest or overlap, there- fore each token can belong to at most one mention; and, (2) mentions comprise continuous sequences of tokens. Nested entity recognition addresses vio- lations of the first assumption (Lu and Roth, 2015; Katiyar and Cardie, 2018; Sohrab and Miwa, 2018; Ringland et al., 2019). However, the violation of The�left�atrium�is�mildly�dilated�. E1 E1 have�much�muscle�pain�and�fatigue�. E2 E3 E3 Figure 1: Examples involving discontinuous mentions, taken from the ShARe 13 (Pradhan et al., 2013) and CADEC (Karimi et al., 2015a) data sets, respectively. The first example contains a discontinuous mention ‘left atrium dilated’, the second example contains two mentions that overlap: ‘muscle pain’ and ‘muscle fa- tigue’ (discontinuous). the second assumption is comparatively less stud- ied and requires handling discontinuous mentions (see examples in Figure 1). In contrast to continuous mentions which are of- ten short spans of text, discontinuous mentions con- sist of components that are separated by intervals. Recognizing discontinuous mentions is particularly challenging as exhaustive enumeration of possible mentions, including discontinuous and overlapping spans, is exponential in sentence length. Existing approaches for discontinuous NER either suffer from high time complexity (McDonald et al., 2005) or ambiguity in translating intermediate represen- tations into mentions (Tang et al., 2013a; Metke- Jimenez and Karimi, 2016; Muis and Lu, 2016). In addition, current art uses traditional approaches that rely on manually designed features, which are tailored to recognize specific entity types. Also, these features usually do not generalize well in different …

Content truncated to 500 words