Neural Token Representations and Negation and Speculation Scope Detection in Biomedical and General Domain Text

Elena Sergeeva, Henghui Zhu, Amir Tahmasebi, Peter Szolovits


Abstract
Since the introduction of context-aware token representation techniques such as Embeddings from Language Models (ELMo) and Bidirectional Encoder Representations from Transformers (BERT), there has been numerous reports on improved performance on a variety of natural language tasks. Nevertheless, the degree to which the resulting context-aware representations encode information about morpho-syntactic properties of the word/token in a sentence remains unclear. In this paper, we investigate the application and impact of state-of-the-art neural token representations for automatic cue-conditional speculation and negation scope detection coupled with the independently computed morpho-syntactic information. Through this work, We establish a new state-of-the-art for the BioScope and NegPar corpus. More importantly, we provide a thorough analysis of neural representations and additional features interactions, cue-representation for conditioning, discuss model behavior on different datasets and address the annotation-induced biases in the learned representations.
Anthology ID:
D19-6221
Volume:
Proceedings of the Tenth International Workshop on Health Text Mining and Information Analysis (LOUHI 2019)
Month:
November
Year:
2019
Address:
Hong Kong
Editors:
Eben Holderness, Antonio Jimeno Yepes, Alberto Lavelli, Anne-Lyse Minard, James Pustejovsky, Fabio Rinaldi
Venue:
Louhi
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
178–187
Language:
URL:
https://aclanthology.org/D19-6221
DOI:
10.18653/v1/D19-6221
Bibkey:
Cite (ACL):
Elena Sergeeva, Henghui Zhu, Amir Tahmasebi, and Peter Szolovits. 2019. Neural Token Representations and Negation and Speculation Scope Detection in Biomedical and General Domain Text. In Proceedings of the Tenth International Workshop on Health Text Mining and Information Analysis (LOUHI 2019), pages 178–187, Hong Kong. Association for Computational Linguistics.
Cite (Informal):
Neural Token Representations and Negation and Speculation Scope Detection in Biomedical and General Domain Text (Sergeeva et al., Louhi 2019)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-bitext-workshop/D19-6221.pdf