A Event Extraction Method of Document-Level Based on the Self-attention Mechanism

Xueming Qiao; Yao Tang; Yanhong Liu; Maomao Su; Chao Wang; Yansheng Fu; Xiaofang Li; Mingrui Wu; Qiang Fu; Dongjie Zhu

doi:10.1007/978-3-031-20099-1_50

A Event Extraction Method of Document-Level Based on the Self-attention Mechanism

Xueming Qiao
, Yao Tang
, Yanhong Liu
, Maomao Su
, Chao Wang
, Yansheng Fu
, Xiaofang Li
, Mingrui Wu
, Qiang Fu
, Dongjie Zhu^*

^*Corresponding author for this work

State Grid Weihai Power Supply Company
School of Computer Science and Technology, Harbin Institute of Technology
Harbin Institute of Technology Weihai
Ltd.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

Event extraction is an important task in the field of natural language processing. However, most of the existing event extraction techniques focus on sentence-level extraction, which inevitably ignores the contextual features of sentences and the occurrence of multiple event trigger words in the same sentence. Therefore, this paper mainly uses the multi-head self-attention mechanism to integrate text features from multiple dimensions and levels to achieve the task of event detection at the level of text. First, convolutional neural network combined with dynamic multi-pool strategy is used to extract sentence level features. Secondly, the discourse feature representation of full-text information fusion is obtained by multi-head self-attention mechanism model. Finally, using the classifier function to classify, and then detect the trigger word and category of the event. Experimental results show that the proposed method achieves good results in document-level event extraction.

Original language	English
Title of host publication	Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings
Editors	Yuan Xu, Hongyang Yan, Huang Teng, Jun Cai, Jin Li
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	609-619
Number of pages	11
ISBN (Print)	9783031200984
DOIs	https://doi.org/10.1007/978-3-031-20099-1_50
State	Published - 2023
Externally published	Yes
Event	4th International Conference on Machine Learning for Cyber Security, ML4CS 2022 - Guangzhou, China Duration: 2 Dec 2022 → 4 Dec 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13656 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	4th International Conference on Machine Learning for Cyber Security, ML4CS 2022
Country/Territory	China
City	Guangzhou
Period	2/12/22 → 4/12/22

Keywords

Document-level
Event detection
Event extraction

Access to Document

10.1007/978-3-031-20099-1_50

Cite this

Qiao, X., Tang, Y., Liu, Y., Su, M., Wang, C., Fu, Y., Li, X., Wu, M., Fu, Q., & Zhu, D. (2023). A Event Extraction Method of Document-Level Based on the Self-attention Mechanism. In Y. Xu, H. Yan, H. Teng, J. Cai, & J. Li (Eds.), Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings (pp. 609-619). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13656 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-20099-1_50

Qiao, Xueming ; Tang, Yao ; Liu, Yanhong et al. / A Event Extraction Method of Document-Level Based on the Self-attention Mechanism. Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings. editor / Yuan Xu ; Hongyang Yan ; Huang Teng ; Jun Cai ; Jin Li. Springer Science and Business Media Deutschland GmbH, 2023. pp. 609-619 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{4274ac71a0654be5a24f64c223b1b29f,

title = "A Event Extraction Method of Document-Level Based on the Self-attention Mechanism",

abstract = "Event extraction is an important task in the field of natural language processing. However, most of the existing event extraction techniques focus on sentence-level extraction, which inevitably ignores the contextual features of sentences and the occurrence of multiple event trigger words in the same sentence. Therefore, this paper mainly uses the multi-head self-attention mechanism to integrate text features from multiple dimensions and levels to achieve the task of event detection at the level of text. First, convolutional neural network combined with dynamic multi-pool strategy is used to extract sentence level features. Secondly, the discourse feature representation of full-text information fusion is obtained by multi-head self-attention mechanism model. Finally, using the classifier function to classify, and then detect the trigger word and category of the event. Experimental results show that the proposed method achieves good results in document-level event extraction.",

keywords = "Document-level, Event detection, Event extraction",

author = "Xueming Qiao and Yao Tang and Yanhong Liu and Maomao Su and Chao Wang and Yansheng Fu and Xiaofang Li and Mingrui Wu and Qiang Fu and Dongjie Zhu",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 4th International Conference on Machine Learning for Cyber Security, ML4CS 2022 ; Conference date: 02-12-2022 Through 04-12-2022",

year = "2023",

doi = "10.1007/978-3-031-20099-1\_50",

language = "英语",

isbn = "9783031200984",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

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

pages = "609--619",

editor = "Yuan Xu and Hongyang Yan and Huang Teng and Jun Cai and Jin Li",

booktitle = "Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings",

address = "德国",

}

Qiao, X, Tang, Y, Liu, Y, Su, M, Wang, C, Fu, Y, Li, X, Wu, M, Fu, Q & Zhu, D 2023, A Event Extraction Method of Document-Level Based on the Self-attention Mechanism. in Y Xu, H Yan, H Teng, J Cai & J Li (eds), Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13656 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 609-619, 4th International Conference on Machine Learning for Cyber Security, ML4CS 2022, Guangzhou, China, 2/12/22. https://doi.org/10.1007/978-3-031-20099-1_50

A Event Extraction Method of Document-Level Based on the Self-attention Mechanism. / Qiao, Xueming; Tang, Yao; Liu, Yanhong et al.
Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings. ed. / Yuan Xu; Hongyang Yan; Huang Teng; Jun Cai; Jin Li. Springer Science and Business Media Deutschland GmbH, 2023. p. 609-619 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13656 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Event Extraction Method of Document-Level Based on the Self-attention Mechanism

AU - Qiao, Xueming

AU - Tang, Yao

AU - Liu, Yanhong

AU - Su, Maomao

AU - Wang, Chao

AU - Fu, Yansheng

AU - Li, Xiaofang

AU - Wu, Mingrui

AU - Fu, Qiang

AU - Zhu, Dongjie

PY - 2023

Y1 - 2023

N2 - Event extraction is an important task in the field of natural language processing. However, most of the existing event extraction techniques focus on sentence-level extraction, which inevitably ignores the contextual features of sentences and the occurrence of multiple event trigger words in the same sentence. Therefore, this paper mainly uses the multi-head self-attention mechanism to integrate text features from multiple dimensions and levels to achieve the task of event detection at the level of text. First, convolutional neural network combined with dynamic multi-pool strategy is used to extract sentence level features. Secondly, the discourse feature representation of full-text information fusion is obtained by multi-head self-attention mechanism model. Finally, using the classifier function to classify, and then detect the trigger word and category of the event. Experimental results show that the proposed method achieves good results in document-level event extraction.

AB - Event extraction is an important task in the field of natural language processing. However, most of the existing event extraction techniques focus on sentence-level extraction, which inevitably ignores the contextual features of sentences and the occurrence of multiple event trigger words in the same sentence. Therefore, this paper mainly uses the multi-head self-attention mechanism to integrate text features from multiple dimensions and levels to achieve the task of event detection at the level of text. First, convolutional neural network combined with dynamic multi-pool strategy is used to extract sentence level features. Secondly, the discourse feature representation of full-text information fusion is obtained by multi-head self-attention mechanism model. Finally, using the classifier function to classify, and then detect the trigger word and category of the event. Experimental results show that the proposed method achieves good results in document-level event extraction.

KW - Document-level

KW - Event detection

KW - Event extraction

UR - https://www.scopus.com/pages/publications/85148694879

U2 - 10.1007/978-3-031-20099-1_50

DO - 10.1007/978-3-031-20099-1_50

M3 - 会议稿件

AN - SCOPUS:85148694879

SN - 9783031200984

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 609

EP - 619

BT - Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings

A2 - Xu, Yuan

A2 - Yan, Hongyang

A2 - Teng, Huang

A2 - Cai, Jun

A2 - Li, Jin

PB - Springer Science and Business Media Deutschland GmbH

T2 - 4th International Conference on Machine Learning for Cyber Security, ML4CS 2022

Y2 - 2 December 2022 through 4 December 2022

ER -

Qiao X, Tang Y, Liu Y, Su M, Wang C, Fu Y et al. A Event Extraction Method of Document-Level Based on the Self-attention Mechanism. In Xu Y, Yan H, Teng H, Cai J, Li J, editors, Machine Learning for Cyber Security - 4th International Conference, ML4CS 2022, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 609-619. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-20099-1_50

A Event Extraction Method of Document-Level Based on the Self-attention Mechanism

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this