Investigating Domain-Specific Information for Neural Coreference Resolution on Biomedical Texts

Abstract

Existing biomedical coreference resolution systems depend on features and/or rules based on syntactic parsers. In this paper, we investigate the utility of the stateof-the-art general domain neural coreference resolution system on biomedical texts. The system is an end-to-end system without depending on any syntactic parsers. We also investigate the domain specific features to enhance the system for biomedical texts. Experimental results on the BioNLP Protein Coreference dataset and the CRAFT corpus show that, with no parser information, the adapted system compared favorably with the systems that depend on parser information on these datasets, achieving 51.23% on the BioNLP dataset and 36.33% on the CRAFT corpus in F1 score. In-domain embeddings and domain-specific features helped improve the performance on the BioNLP dataset, but they did not on the CRAFT corpus.

Context-Aware Feature Attention Model for Coreference Resolution on Biomedical Texts

“…To further understand the utility of the Feature Atten-2 tion mechanism for mention detection subtask, we list 3 the mention detection performance in Table 5. Over-4 all, compared with [9], the performance of the proposed SFA model is significantly increased by 6.3 F1 on shown in Figure 7. For the three "it" in the figure , we can find that the first two "it" that are coreferential gain similar weights for the there features, where the weight of span width is the highest, followed by the grammatical number, and finally Metamap.…”

“…1) rule and feature-based models [4,5,6], which heavily rely on syntactic parsers to extract manually crafted features and rules. 2) hybrid models [7,8], which combine rule-based and machine learning-based methods for biomedical coreference resolution 3) neural networkbased models [6,9,10], which use deep learning or neural networks to solve the problem automatically with domain-specific information integration, typically by 1 pre-trained embeddings and some biomedical features.…”

“…Our work is most closely related to the work of [9], 4 while we focus on the problem that identical mentions 5 tend to get similar or even identical representations, 6 which further mislead to make coreferential mistakes. In an end-to-end coreference resolution system, the in-10 put is a document D with T words, and the output is 11 a set of mention clusters.…”

Named Entity Recognition and Relation Detection for Biomedical Information Extraction

“…Recently, deep learning methods have been used for coreference resolution in general domain successfully without using syntactic parsers, for example in Lee et al ( 2017 ). The same system has been applied to biomedical coreference resolution in Trieu et al ( 2018 ) with some domain-specific feature enhancements. Here, it is worth mentioning that the CRAFT corpus, earlier mentioned in Table 1 , has an improved version that can be used for coreference resolution for biomedical texts (Cohen et al, 2017 ).…”

Coreference Resolution in Full Text Articles with BERT and Syntax-based Mention Filtering

“…• E2E MetaMap (Trieu et al, 2018): this model is based on the baseline model (Lee et al, 2017) but it particularly incorporated semantic type features extracted from the MetaMapLite (Demner-Fushman et al, 2017) to address biomedical documents. The maximum antecedent was 250.…”