This paper presents a novel transfer multi-task learning method for Bacteria Biotope rel+ner task at BioNLP-OST 2019. To alleviate the data deficiency problem in domain-specific information extraction, we use BERT(Devlin et al., 2018) (Bidirectional Encoder Representations from Transformers) and pre-train it using mask language models and next sentence prediction (Devlin et al., 2018) on both general corpus and medical corpus like PubMed. In fine-tuning stage, we fine-tune the relation extraction layer and mention recognition layer designed by us on the top of BERT to extract mentions and relations simultaneously. The evaluation results show that our method achieves the best performance on all metrics (including slot error rate, precision and recall) in the Bacteria Biotope rel+ner subtask. c 2019 Association for Computational Linguistics.
CITATION STYLE
Zhang, Q., Liu, C., Chi, Y., Xie, X., & Hua, X. (2019). A Multi-Task Learning Framework for Extracting Bacteria Biotope Information. In BioNLP-OST@EMNLP-IJNCLP 2019 - Proceedings of the 5th Workshop on BioNLP Open Shared Tasks (pp. 105–109). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/d19-5716
Mendeley helps you to discover research relevant for your work.