Joint event extraction aims at discovering event from texts and simultaneously identifying their corresponding event types, argument roles, which is an essential but challenging task in natural language processing. Although extensively studied, existing approaches still suffered from low accuracy. To solve this issue, this paper proposes a deep learning-based joint event extraction approach, JBGCN-MATT, which applies deep pre-trained language model to represent texts and combines syntactic graph convolution network and multi-attention mechanism to capture long-distance dependencies. Comprehensive experiments were conducted on the benchmark dataset ACE2005, results show that JBGCN-MATT achieves the F1 score of 74.2% and 60.5% for the trigger classification task and the argument classification task, respectively, and outperforms the state-of-the-art methods.
LiQLiJXShengJW, et al.A survey on deep learning event extraction: approaches and applications. IEEE Transact Neural Networks Learn Syst2024; 35(5): 6301–6321.
2.
XiangWWangB. A survey of event extraction from text. IEEE Access2019; 7: 173111–173137.
3.
AtriYKGoyalVChakrabortyT. Multi-document summarization using selective attention span and reinforcement learning. IEEE/ACM Trans Audio Speech Lang Process2023; 31: 3457–3467.
4.
NazirARaoYWuLW, et al.Issues and challenges of aspect-based sentiment analysis: a comprehensive survey. IEEE Trans Affect Comput2022; 13(2): 845–863.
5.
HuangHYLiuXShiG, et al.Event extraction with dynamic prefix tuning and relevance retrieval. IEEE Trans Knowl Data Eng2023; 35(10): 9946–9958.
6.
SongDDXuJPangJH, et al.Classifier-adaptation knowledge distillation framework for relation extraction and event detection with imbalanced data. Inf Sci2021; 573: 222–238.
7.
LuXSZhouMCWuKY. A novel fuzzy logic-based text classification method for tracking rare events on twitter. IEEE Trans Syst Man Cybern Syst2021; 51(7): 4324–4333.
8.
ChengQFuYHHuangJC, et al.Event detection based on the label attention mechanism. Int J Mach Learn Cybern2023; 14(2): 633–641.
9.
HuBLiuYChenNY, et al.SEGCN-DCR: a syntax-enhanced event detection framework with decoupled classification rebalance. Neurocomputing2022; 481: 55–66.
10.
BalaliAAsadpourMCamposR, et al.Joint event extraction along shortest dependency paths using graph convolutional networks. Knowl Base Syst2021; 210: 106492.
11.
ZhangJCHeQZhangY. Syntax grounded graph convolutional network for joint entity and event extraction. Neurocomputing2021; 422: 118–128.
12.
El-allalyEDSarroutiMEn-NahnahiN, et al.An attentive joint model with transformer-based weighted graph convolutional network for extracting adverse drug event relation. J Biomed Inf2022; 125: 103968.
13.
CekinelRFKaragozP. Event prediction from news text using subgraph embedding and graph sequence mining. World Wide Web2022; 25(6): 2403–2428.
14.
ChangCTangYLongYX, et al.Multi-information preprocessing event extraction with BiLSTM-CRF attention for academic knowledge graph construction. IEEE Trans Comput Soc Syst2023; 10(5): 2713–2724.
15.
LiDYYanLMaZM. Dependency-based BERT for Chinese event argument extraction. ACM Trans Asian Low-Resour Lang Inf Process2023; 22(12): 259.
16.
ChenQYangKHGuoX, et al.Joint overlapping event extraction model via role pre-judgment with trigger and context embeddings. Electronics2023; 12(22): 4688.
17.
HuangHChenYPLCHuangRZ, et al.A multi-graph representation for event extraction. Artif Intell2024; 332: 104144.
18.
YangJGaoHDangDP. Graph convolutional networks with syntactic and semantic structures for event detection. IEEE Access2024; 12: 64949–64957.
19.
LiuZLiYMZhangY, et al.Effective event extraction method via enhanced graph convolutional network indication with hierarchical argument selection strategy. Electronics2023; 12(13): 2981.
20.
HeXYYanGSiCF, et al.General fine-grained event detection based on fusion of multi-information representation and attention mechanism. Int J Mach Learn Cybern2023; 14(12): 4393–4403.
21.
HaffarNZriguiM. A synergistic bidirectional LSTM and N-gram multi-channel CNN Approach based on BERT and FastText for Arabic event identification. ACM Trans Asian Low-Resour Lang Inf Process2023; 22(11): 246.
22.
DingLChenXJWeiJ, et al.MABERT: mask-attention-based BERT for Chinese event extraction. ACM Trans Asian Low-Resour Lang Inf Process2023; 22(7): 192.
23.
YuXYRongWGLiuJS, et al.LSTM-based end-to-end framework for biomedical event extraction. IEEE ACM Trans Comput Biol Bioinf2020; 17(6): 2029–2039.
24.
WangJWangKXYaoY, et al.Event detection with multi-order edge-aware graph convolution networks. Data Knowl Eng2023; 143: 102109.
