A Span-Based Model for Joint Overlapped and Discontinuous Named Entity Recognition

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 4814–4828 August 1–6, 2021. ©2021 Association for Computational Linguistics 4814 A Span-Based Model for Joint Overlapped and Discontinuous Named Entity Recognition Fei Li1 and Zhichao Lin2 and Meishan Zhang2 and Donghong Ji1∗ 1. Department of Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, China 2. School of New Media and Communication, Tianjin University, China {lifei csnlp,dhji}@whu.edu.cn and mason.zms@gmail.com Abstract Research on overlapped and discontinuous named entity recognition (NER) has received increasing attention. The majority of previ- ous work focuses on either overlapped or dis- continuous entities. In this paper, we pro- pose a novel span-based model that can rec- ognize both overlapped and discontinuous en- tities jointly. The model includes two ma- jor steps. First, entity fragments are recog- nized by traversing over all possible text spans, thus, overlapped entities can be recognized. Second, we perform relation classification to judge whether a given pair of entity fragments to be overlapping or succession. In this way, we can recognize not only discontinuous en- tities, and meanwhile doubly check the over- lapped entities. As a whole, our model can be regarded as a relation extraction paradigm essentially. Experimental results on multiple benchmark datasets (i.e., CLEF, GENIA and ACE05) show that our model is highly compet- itive for overlapped and discontinuous NER. 1 Introduction Named entity recognition (NER) (Sang and De Meulder, 2003) is one fundamental task for natural language processing (NLP), due to its wide application in information extraction and data min- ing (Lin et al., 2019b; Cao et al., 2019). Tradi- tionally, NER is presented as a sequence labeling problem and widely solved by conditional random field (CRF) based models (Lafferty et al., 2001). However, this framework is difficult to handle over- lapped and discontinuous entities (Lu and Roth, 2015; Muis and Lu, 2016), which we illustrate us- ing two examples as shown in Figure 1. The two entities “Pennsylvania” and “Pennsylvania radio station” are nested with each other,1 and the sec- ∗Corresponding author. 1 We consider “nested” as a special case of “overlapped”. ond example shows a discontinuous entity “mitral leaflets thickened” involving three fragments. There have been several studies to investigate overlapped or discontinuous entities (Finkel and Manning, 2009; Lu and Roth, 2015; Muis and Lu, 2017; Katiyar and Cardie, 2018; Wang and Lu, 2018; Ju et al., 2018; Wang et al., 2018; Fisher and Vlachos, 2019; Luan et al., 2019; Wang and Lu, 2019). The majority of them focus on over- lapped NER, with only several exceptions to the best of our knowledge. Muis and Lu (2016) present a hypergraph model that is capable of handling both overlapped and discontinuous entities. Wang and Lu (2019) extend the hypergraph model with long short-term memories (LSTMs) (Hochreiter and Schmidhuber, 1997). Dai et al. (2020) pro- posed a transition-based neural model for discon- tinuous NER. By …

Content truncated to 500 words