Similar articles

Abstract: event extraction (EE) is a complex natural language processing (NLP) task that aims at identifying and classifying triggers and arguments in raw text. The polysemy of triggers and arguments stands out as one of the key challenges affecting the precise extraction of events. Existing approaches commonly consider the semantic distribution of triggers and arguments to be balanced. However, the sample quantities of different semantics in the same trigger or argument vary in real-world scenarios, leading to a biased semantic distribution. The bias introduces two challenges: (1) low-frequency semantics is difficult to identify; (2) high-frequency semantics is often mistakenly identified. To tackle these challenges, we propose an adaptive learning method with the reward–;penalty mechanism for balancing the semantic distribution in polysemous triggers and arguments. The reward–;penalty mechanism balances the semantic distribution by enlarging the gap between the target and nontarget semantics by rewarding correct classifications and penalizing incorrect classifications. Additionally, we propose a sentence-level event situation awareness (SA) mechanism to guide the encoder to accurately learn the knowledge of events mentioned in the sentence, thereby enhancing target event semantics in the distribution of polysemous triggers and arguments. Finally, for various semantics in different tasks, we propose task-specific semantic decoders to precisely identify the boundaries of the predicted triggers and arguments for the semantics. Our experimental results on ACE2005 and its variants, along with the rich Entities, Relations, and Events (ERE), demonstrate the superiority of our approach over single-task and multi-task EE baselines.

[1]Ahn D, 2006. The stages of event extraction. Proc Workshop on Annotating and Reasoning about Time and Events, p.1-8.

[2]Akgun SA, Ghafurian M, Crowley M, et al., 2023. Using affect as a communication modality to improve human-robot communication in robot-assisted search and rescue scenarios. IEEE Trans Affect Comput, 14(4):3013-3030.

[3]Anelli VW, Di Noia T, Di Sciascio E, et al., 2022. Inferring user decision-making processes in recommender systems with knowledge graphs. Proc 30^th Italian Symp on Advanced Database Systems, p.505-513.

[4]Chen YB, Xu LH, Liu K, et al., 2015. Event extraction via dynamic multi-pooling convolutional neural networks. Proc 53^rd Annual Meeting of the Association for Computational Linguistics and 7^th Int Joint Conf on Natural Language Processing, p.167-176.

[5]Cong X, Cui SY, Yu BW, et al., 2021. Few-shot event detection with prototypical amortized conditional random field. Findings of the Association for Computational Linguistics, p.28-40.

[6]Cui SY, Yu BW, Liu TW, et al., 2020. Edge-enhanced graph convolution networks for event detection with syntactic relation. Findings of the Association for Computational Linguistics, p.2329-2339.

[7]Cui ZJ, Yuan ZM, Wu YF, et al., 2023. Intelligent recommendation for departments based on medical knowledge graph. IEEE Access, 11:25372-25385.

[8]Devlin J, Chang MW, Lee K, et al., 2019. BERT: pre-training of deep bidirectional Transformers for language understanding. Proc Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, p.4171-4186.

[9]Ding N, Li ZR, Liu ZY, et al., 2019. Event detection with trigger-aware lattice neural network. Proc Conf on Empirical Methods in Natural Language Processing and 9^th Int Joint Conf on Natural Language Processing, p.347-356.

[10]Dozat T, Manning CD, 2017. Deep biaffine attention for neural dependency parsing. Proc 5^th Int Conf on Learning Representations, p.1-8.

[11]Du XY, Cardie C, 2020. Event extraction by answering (almost) natural questions. Proc Conf on Empirical Methods in Natural Language Processing, p.671-683.

[12]Du XY, Ji H, 2022. Retrieval-augmented generative question answering for event argument extraction. Proc Conf on Empirical Methods in Natural Language Processing, p.4649-4666.

[13]Dwivedi D, Yemula PK, Pal M, 2023. DynamoPMU: a physics informed anomaly detection, clustering, and prediction method using nonlinear dynamics on μ PMU measurements. IEEE Trans Instrum Meas, 72:1-9.

[14]Endsley MR, 1988. Design and evaluation for situation awareness enhancement. Proc Hum Factors Ergon Soc Ann Meet, 32(2):97-101.

[15]Endsley MR, 2001. Designing for situation awareness in complex systems. Proc 2^nd Int Workshop on Symbiosis of Humans, Artifacts and Environment, p.1-14.

[16]Ettinger A, Hwang J, Pyatkin V, et al., 2023. “You are an expert linguistic annotator”: limits of LLMs as analyzers of abstract meaning representation. Findings of the Association for Computational Linguistics, p.8250-8263.

[17]Feng XC, Qin B, Liu T, 2018. A language-independent neural network for event detection. Sci China Inform Sci, 61(9):092106.

[18]Gururangan S, Marasović A, Swayamdipta S, et al., 2020. Don’t stop pretraining: adapt language models to domains and tasks. Proc 58^th Annual Meeting of the Association for Computational Linguistics, p.8342-8360.

[19]Guzman-Nateras L, Dernoncourt F, Nguyen T, 2023. Hybrid knowledge transfer for improved cross-lingual event detection via hierarchical sample selection. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.5414-5427.

[20]Han RD, Yang CH, Peng T, et al., 2023. An empirical study on information extraction using large language models.

[21]He YX, Hu JY, Tang BZ, 2023. Revisiting event argument extraction: can EAE models learn better when being aware of event co-occurrences? Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.12542-12556.

[22]Hong Y, Zhang JF, Ma B, et al., 2011. Using cross-entity inference to improve event extraction. Proc 49^th Annual Meeting of the Association for Computational Linguistics, p.1127-1136.

[23]Hsu IH, Huang KH, Boschee E, et al., 2022. DEGREE: a data-efficient generation-based event extraction model. Proc Conf of the North American Chapter of the Association for Computational Linguistics, p.1890-1908.

[24]Hsu IH, Xie ZY, Huang KH, et al., 2023. AMPERE: AMR-aware prefix for generation-based event argument extraction model. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.10976-10993.

[25]Ji H, Grishman R, 2008. Refining event extraction through cross-document inference. Association for Computational Linguistics Annual Meeting: Human Language Technologies, p.254-262.

[26]Lewis M, Liu YH, Goyal N, et al., 2020. BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. Proc 58^th Annual Meeting of the Association for Computational Linguistics, p.7871-7880.

[27]Li S, Zhao RN, Li ML, et al., 2023. Open-domain hierarchical event schema induction by incremental prompting and verification. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.5677-5697.

[28]Lin Y, Ji H, Huang F, et al., 2020. A joint neural model for information extraction with global features. Proc 58^th Annual Meeting of the Association for Computational Linguistics, p.7999-8009.

[29]Liu J, Chen YF, Xu JA, 2021. Machine reading comprehension as data augmentation: a case study on implicit event argument extraction. Proc Conf on Empirical Methods in Natural Language Processing, p.2716-2725.

[30]Liu J, Chen YF, Xu JA, 2022. Saliency as evidence: event detection with trigger saliency attribution. Proc 60^th Annual Meeting of the Association for Computational Linguistics, p.4573-4585.

[31]Liu J, Sui DB, Liu K, et al., 2023. Learning with partial annotations for event detection. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.508-523.

[32]Liu X, Luo ZC, Huang HY, 2018. Jointly multiple events extraction via attention-based graph information aggregation. Proc Conf on Empirical Methods in Natural Language Processing, p.1247-1256.

[33]Liu X, Huang HY, Shi G, et al., 2022. Dynamic prefix-tuning for generative template-based event extraction. Proc 60^th Annual Meeting of the Association for Computational Linguistics, p.5216-5228.

[34]Loshchilov I, Hutter F, 2019. Decoupled weight decay regularization. Proc 7^th Int Conf on Learning Representations, p.1-8.

[35]Lou DF, Liao ZL, Deng SM, et al., 2021. MLBiNet: a cross-sentence collective event detection network. Proc 59^th Annual Meeting of the Association for Computational Linguistics and 11^th Int Joint Conf on Natural Language Processing, p.4829-4839.

[36]Lu D, Ran SH, Tetreault JR, et al., 2023. Event extraction as question generation and answering. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.1666-1688.

[37]Lu YJ, Lin HY, Xu J, et al., 2021. Text2Event: controllable sequence-to-structure generation for end-to-end event extraction. Proc 59^th Annual Meeting of the Association for Computational Linguistics and 11^th Int Joint Conf on Natural Language Processing, p.2795-2806.

[38]Ma MD, Taylor A, Wang W, et al., 2023. DICE: data-efficient clinical event extraction with generative models. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.15898-15917.

[39]Matey AH, Danquah P, Koi-Akrofi GY, 2022. Predicting cyber-attack using cyber situational awareness: the case of independent power producers (IPPs). Int J Adv Comput Sci Appl, 13(1):700-709.

[40]McClosky D, Surdeanu M, Manning CD, 2011. Event extraction as dependency parsing. Proc 49^th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, p.1626-1635.

[41]Nam H, Kim SH, Park YH, 2022. FilterAugment: an acoustic environmental data augmentation method. Proc IEEE Int Conf on Acoustics, Speech and Signal Processing, p.4308-4312.

[42]Nguyen TH, Cho K, Grishman R, 2016. Joint event extraction via recurrent neural networks. Proc Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, p.300-309.

[43]Onwubiko C, 2020. CyberOps: situational awareness in cybersecurity operations. Int J Cyber Situat Aware, 5(1):82-107.

[44]Palash M, Bhargava B, 2023. SAFER: situation aware facial emotion recognition.

[45]Pang CX, Cao YX, Ding Q, et al., 2023. Guideline learning for in-context information extraction. Proc Conf on Empirical Methods in Natural Language Processing, p.15372-15389.

[46]Ping Y, Lu JY, Gan RY, et al., 2023. UniEX: an effective and efficient framework for unified information extraction via a span-extractive perspective. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.16424-16440.

[47]Sha L, Qian F, Chang BB, et al., 2018. Jointly extracting event triggers and arguments by dependency-bridge RNN and tensor-based argument interaction. Proc 32^nd AAAI Conf on Artificial Intelligence, p.5916-5923.

[48]Shashikumar SP, Josef CS, Sharma A, et al., 2021. DeepAISE—an interpretable and recurrent neural survival model for early prediction of sepsis. Artif Intell Med, 113:102036.

[49]Shu XF, Yan J, Gao WR, et al., 2021. Research on military equipment entity recognition and knowledge graph construction method based on ALBERT-Bi-LSTM-CRF. Proc 4^th Int Conf on Artificial Intelligence and Pattern Recognition, p.273-279.

[50]Song ZY, Bies A, Strassel S, et al., 2015. From light to rich ERE: annotation of entities, relations, and events. Proc 3^rd Workshop on EVENTS: Definition, Detection, Coreference, and Representation, p.89-98.

[51]Wadden D, Wennberg U, Luan Y, et al., 2019. Entity, relation, and event extraction with contextualized span representations. Proc Conf on Empirical Methods in Natural Language Processing and 9^th Int Joint Conf on Natural Language Processing, p.5783-5788.

[52]Wang B, Huang HY, Wei XC, et al., 2023. Boosting event extraction with denoised structure-to-text augmentation. Findings of the Association for Computational Linguistics, p.11267-11281.

[53]Wang S, Yu M, Chang S, et al., 2022. Query and extract: refining event extraction as type-oriented binary decoding. Findings of the Association for Computational Linguistics, p.169-182.

[54]Wang SJ, Yu M, Huang LF, 2023. The art of prompting: event detection based on type specific prompts. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.1286-1299.

[55]Wang ZT, Wang XY, Hu W, 2023. Continual event extraction with semantic confusion rectification. Proc Conf on Empirical Methods in Natural Language Processing, p.11945-11955.

[56]Xia LQ, Liang YS, Leng JW, et al., 2023. Maintenance planning recommendation of complex industrial equipment based on knowledge graph and graph neural network. Reliab Eng Syst Saf, 232:109068.

[57]Xu RX, Wang PY, Liu TY, et al., 2022. A two-stream AMR-enhanced model for document-level event argument extraction. Proc Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, p.5025-5036.

[58]Xu TY, Guo C, Du LX, et al., 2022. A method for traditional Chinese medicine knowledge graph dynamic construction. Proc 5^th Int Conf on Big Data Technologies, p.196-202.

[59]Xu ZY, Lee JY, Huang LF, 2023. Learning from a friend: improving event extraction via self-training with feedback from abstract meaning representation. Findings of the Association for Computational Linguistics, p.10421-10437.

[60]Xu ZZ, Liu RK, Yang S, et al., 2023. Learning imbalanced data with vision transformers. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.15793-15803.

[61]Yang P, Cong X, Sun ZY, et al., 2021. Enhanced language representation with label knowledge for span extraction. Proc Conf on Empirical Methods in Natural Language Processing, p.4623-4635.

[62]Yang S, Feng DW, Qiao LB, et al., 2019. Exploring pre-trained language models for event extraction and generation. Proc 57^th Annual Meeting of the Association for Computational Linguistics, p.5284-5294.

[63]Yang XJ, Lu YJ, Petzold LR, 2023. Few-shot document-level event argument extraction. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.8029-8046.

[64]Yang YQ, Guo QP, Hu XK, et al., 2023. An AMR-based link prediction approach for document-level event argument extraction. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.12876-12889.

[65]Yao YZ, Mao SY, Zhang NY, et al., 2023. Schema-aware reference as prompt improves data-efficient knowledge graph construction. Proc 46^th Int ACM SIGIR Conf on Research and Development in Information Retrieval, p.911-921.

[66]You HL, Samuel D, Touileb S, et al., 2022. EventGraph: event extraction as semantic graph parsing. Proc 5^th Workshop on Challenges and Applications of Automated Extraction of Socio-Political Events from Text, p.7-15.

[67]You HL, Touileb S, Øvrelid L, 2023. JSEEGraph: joint structured event extraction as graph parsing. Proc 12^th Joint Conf on Lexical and Computational Semantics, p.115-127.

[68]You MS, Yin J, Wang H, et al., 2023. A knowledge graph empowered online learning framework for access control decision-making. World Wide Web, 26(2):827-848.

[69]Zeng Y, Yang HH, Feng YS, et al., 2016. A convolution BiLSTM neural network model for Chinese event extraction. Proc 5^th CCF Conf on Natural Language Processing and Chinese Computing, 10102:275-287.

[70]Zhang JR, Ilievski F, Ma KX, et al., 2023. A study of situational reasoning for traffic understanding. Proc 29^th ACM SIGKDD Conf on Knowledge Discovery and Data Mining, p.3262-3272.

[71]Zhang KH, Shuang K, Yang XY, et al., 2023. What is overlap knowledge in event argument extraction? APE: a cross-datasets transfer learning model for EAE. Proc 61^st Annual Meeting of the Association for Computational Linguistics, p.393-409.

[72]Zhao G, Gong XC, Yang XJ, et al., 2023. DemoSG: demonstration-enhanced schema-guided generation for low-resource event extraction. Findings of the Association for Computational Linguistics, p.1805-1816.

[73]Zheng YZ, Pan SR, Lee VCS, et al., 2022. Rethinking and scaling up graph contrastive learning: an extremely efficient approach with group discrimination. Conf on Neural Information Processing Systems, p.1-17.