PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain

Yuehu Dong; Dongmei Li; Jinghang Gu; Longhua Qian; Guodong Zhou

doi:10.1007/978-981-99-9864-7_2

PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain

Yuehu Dong, Dongmei Li, Jinghang Gu, Longhua Qian (Corresponding Author), Guodong Zhou

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

In this paper, we propose a simple and effective few-shot named entity recognition (NER) method for biomedical domain, called PEMRC (Positive Enhanced Machine Reading Comprehension). PEMRC is based on the idea of using machine reading comprehension reading comprehension (MRC) framework to perfome few-shot NER and fully exploit the prior knowledge implied in the label information. On one hand, we design three different query templates to better induce knowledge from pre-trained language models (PLMs). On the other hand, we design a positive enhanced loss function to improve the model’s accuracy in identifying the start and end positions of entities under low-resources scenarios. Extensive experimental results on eight benchmark datasets in biomedical domain show that PEMRC significantly improves the performance of few-shot NER.

Original language	English
Title of host publication	Health Information Processing
Subtitle of host publication	9th China Health Information Processing Conference, CHIP 2023, Proceedings
Editors	Hua Xu, Qingcai Chen, Hongfei Lin, Fei Wu, Lei Liu, Buzhou Tang, Tianyong Hao, Zhengxing Huang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	22-35
Number of pages	14
ISBN (Print)	9789819998630
DOIs	https://doi.org/10.1007/978-981-99-9864-7_2
Publication status	Published - 1 Feb 2024
Event	9th China Health Information Processing Conference, CHIP 2023 - Hangzhou, China Duration: 27 Oct 2023 → 29 Oct 2023

Publication series

Name	Communications in Computer and Information Science
Volume	1993 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	9th China Health Information Processing Conference, CHIP 2023
Country/Territory	China
City	Hangzhou
Period	27/10/23 → 29/10/23

Keywords

Biomedical Domain
Few-shot Named Entity Recognition
Machine Reading Comprehension

ASJC Scopus subject areas

General Computer Science
General Mathematics

More information

10.1007/978-981-99-9864-7_2

http://hdl.handle.net/10397/107980

Cite this

Dong, Y., Li, D., Gu, J., Qian, L., & Zhou, G. (2024). PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain. In H. Xu, Q. Chen, H. Lin, F. Wu, L. Liu, B. Tang, T. Hao, & Z. Huang (Eds.), Health Information Processing: 9th China Health Information Processing Conference, CHIP 2023, Proceedings (pp. 22-35). (Communications in Computer and Information Science; Vol. 1993 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-99-9864-7_2

Dong, Yuehu ; Li, Dongmei ; Gu, Jinghang et al. / PEMRC : A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain. Health Information Processing: 9th China Health Information Processing Conference, CHIP 2023, Proceedings. editor / Hua Xu ; Qingcai Chen ; Hongfei Lin ; Fei Wu ; Lei Liu ; Buzhou Tang ; Tianyong Hao ; Zhengxing Huang. Springer Science and Business Media Deutschland GmbH, 2024. pp. 22-35 (Communications in Computer and Information Science).

@inproceedings{ce2fd8b8eac445a6a4147ccb8778db5a,

title = "PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain",

abstract = "In this paper, we propose a simple and effective few-shot named entity recognition (NER) method for biomedical domain, called PEMRC (Positive Enhanced Machine Reading Comprehension). PEMRC is based on the idea of using machine reading comprehension reading comprehension (MRC) framework to perfome few-shot NER and fully exploit the prior knowledge implied in the label information. On one hand, we design three different query templates to better induce knowledge from pre-trained language models (PLMs). On the other hand, we design a positive enhanced loss function to improve the model{\textquoteright}s accuracy in identifying the start and end positions of entities under low-resources scenarios. Extensive experimental results on eight benchmark datasets in biomedical domain show that PEMRC significantly improves the performance of few-shot NER.",

keywords = "Biomedical Domain, Few-shot Named Entity Recognition, Machine Reading Comprehension",

author = "Yuehu Dong and Dongmei Li and Jinghang Gu and Longhua Qian and Guodong Zhou",

note = "Publisher Copyright: {\textcopyright} 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 9th China Health Information Processing Conference, CHIP 2023 ; Conference date: 27-10-2023 Through 29-10-2023",

year = "2024",

month = feb,

day = "1",

doi = "10.1007/978-981-99-9864-7_2",

language = "English",

isbn = "9789819998630",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "22--35",

editor = "Hua Xu and Qingcai Chen and Hongfei Lin and Fei Wu and Lei Liu and Buzhou Tang and Tianyong Hao and Zhengxing Huang",

booktitle = "Health Information Processing",

address = "Germany",

}

Dong, Y, Li, D, Gu, J, Qian, L & Zhou, G 2024, PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain. in H Xu, Q Chen, H Lin, F Wu, L Liu, B Tang, T Hao & Z Huang (eds), Health Information Processing: 9th China Health Information Processing Conference, CHIP 2023, Proceedings. Communications in Computer and Information Science, vol. 1993 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 22-35, 9th China Health Information Processing Conference, CHIP 2023, Hangzhou, China, 27/10/23. https://doi.org/10.1007/978-981-99-9864-7_2

PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain. / Dong, Yuehu; Li, Dongmei; Gu, Jinghang et al.
Health Information Processing: 9th China Health Information Processing Conference, CHIP 2023, Proceedings. ed. / Hua Xu; Qingcai Chen; Hongfei Lin; Fei Wu; Lei Liu; Buzhou Tang; Tianyong Hao; Zhengxing Huang. Springer Science and Business Media Deutschland GmbH, 2024. p. 22-35 (Communications in Computer and Information Science; Vol. 1993 CCIS).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - PEMRC

T2 - 9th China Health Information Processing Conference, CHIP 2023

AU - Dong, Yuehu

AU - Li, Dongmei

AU - Gu, Jinghang

AU - Qian, Longhua

AU - Zhou, Guodong

PY - 2024/2/1

Y1 - 2024/2/1

N2 - In this paper, we propose a simple and effective few-shot named entity recognition (NER) method for biomedical domain, called PEMRC (Positive Enhanced Machine Reading Comprehension). PEMRC is based on the idea of using machine reading comprehension reading comprehension (MRC) framework to perfome few-shot NER and fully exploit the prior knowledge implied in the label information. On one hand, we design three different query templates to better induce knowledge from pre-trained language models (PLMs). On the other hand, we design a positive enhanced loss function to improve the model’s accuracy in identifying the start and end positions of entities under low-resources scenarios. Extensive experimental results on eight benchmark datasets in biomedical domain show that PEMRC significantly improves the performance of few-shot NER.

AB - In this paper, we propose a simple and effective few-shot named entity recognition (NER) method for biomedical domain, called PEMRC (Positive Enhanced Machine Reading Comprehension). PEMRC is based on the idea of using machine reading comprehension reading comprehension (MRC) framework to perfome few-shot NER and fully exploit the prior knowledge implied in the label information. On one hand, we design three different query templates to better induce knowledge from pre-trained language models (PLMs). On the other hand, we design a positive enhanced loss function to improve the model’s accuracy in identifying the start and end positions of entities under low-resources scenarios. Extensive experimental results on eight benchmark datasets in biomedical domain show that PEMRC significantly improves the performance of few-shot NER.

KW - Biomedical Domain

KW - Few-shot Named Entity Recognition

KW - Machine Reading Comprehension

UR - http://www.scopus.com/inward/record.url?scp=85186730787&partnerID=8YFLogxK

U2 - 10.1007/978-981-99-9864-7_2

DO - 10.1007/978-981-99-9864-7_2

M3 - Conference article published in proceeding or book

AN - SCOPUS:85186730787

SN - 9789819998630

T3 - Communications in Computer and Information Science

SP - 22

EP - 35

BT - Health Information Processing

A2 - Xu, Hua

A2 - Chen, Qingcai

A2 - Lin, Hongfei

A2 - Wu, Fei

A2 - Liu, Lei

A2 - Tang, Buzhou

A2 - Hao, Tianyong

A2 - Huang, Zhengxing

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 27 October 2023 through 29 October 2023

ER -

Dong Y, Li D, Gu J, Qian L, Zhou G. PEMRC: A Positive Enhanced Machine Reading Comprehension Method for Few-Shot Named Entity Recognition in Biomedical Domain. In Xu H, Chen Q, Lin H, Wu F, Liu L, Tang B, Hao T, Huang Z, editors, Health Information Processing: 9th China Health Information Processing Conference, CHIP 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2024. p. 22-35. (Communications in Computer and Information Science). doi: 10.1007/978-981-99-9864-7_2