封闭域深度学习事件抽取方法研究综述

doi:10.3778/j.issn.1673-9418.2208010

摘要/Abstract

摘要： 事件抽取作为自然语言处理的研究热点之一，其可以从海量数据中提取出有价值的信息，并以结构化的形式展现给用户，在舆情监控、信息检索、金融风投等方面有着广阔的应用前景以及巨大的应用价值。随着深度学习的发展，基于深度学习的事件抽取方法受到越来越多研究人员的关注。通过事件抽取的任务定义，描述基于深度学习事件抽取现有的研究理论；通过分析不同神经网络模型构成的事件抽取方法，探究基于深度学习事件抽取的处理思路，并根据不同的处理思路，将深度学习的事件抽取方法进行分类。在每个分类中，对最经典的研究工作从其问题来源、解决方案、贡献以及缺陷不足进行详细的分析；通过事件抽取数据集的介绍，说明当前事件抽取的研究基础；通过在不同数据集上的事件抽取性能表现，归纳当前事件抽取的发展现状。在此基础上，对当前深度学习应用于事件抽取所面临的困难和挑战进行总结与分析，并对事件抽取的发展前景进行展望。

关键词: 事件抽取, 深度学习, 封闭域, 事件数据集, 自然语言处理

Abstract: As one of the research hotspots in natural language processing, event extraction can extract valuable information from massive data and present it to users in a structured form, which has a broad application prospect as well as great application value in public opinion monitoring, information retrieval, financial venture capital and so on. With the development of deep learning, event extraction methods based on deep learning have received more and more attention from researchers. Through the task definition of event extraction, the existing research theories of deep learning-based event extraction are described; the processing ideas of deep learning-based event extraction are explored by analyzing the event extraction methods composed of different neural network models, and the deep learning-based event extraction methods are classified according to the different processing ideas. In each classification, the most classical research works are analyzed in detail in terms of their problem sources, solutions, contributions and shortcomings; the current research basis of event extraction is illustrated through the introduction of event extraction datasets; the current development status of event extraction is summarized through the performance performance of event extraction on different datasets. On this basis, we summarize and analyze the difficulties and challenges faced by the current deep learning application to event extraction, and provide an outlook on the development prospect of event extraction.

Key words: event extraction, deep learning, closed-domain, event data set, Natural Language Processing(NLP)

焦磊, 云静, 刘利民, 郑博飞, 袁静姝. 封闭域深度学习事件抽取方法研究综述[J]. 计算机科学与探索, 2023, 17(3): 533-548.

JIAO Lei, YUN Jing, LIU Limin, ZHENG Bofei, YUAN Jingshu. Overview of Closed-Domain Deep Learning Event Extraction Methods[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(3): 533-548.

参考文献

[1] 高强, 游宏梁. 事件抽取技术研究综述[J]. 情报理论与实践, 2013, 36(4): 114-117.
GAO Q, YOU H L. A review of research on event extrac-tion techniques[J]. Intelligence Theory and Practice, 2013, 36(4): 114-117.
[2] ROSPOCHER M, VAN E M, VOSSEN P, et al. Building event-centric knowledge graphs from news[J]. Journal of Web Semantics, 2016, 37: 132-151.
[3] LI Z, DING X, LIU T. Constructing narrative event evolu-tionary graph for script event prediction[J]. arXiv:1805.05081, 2018.
[4] CONLON S J, ABRAHAMS A S, SIMMONS L L. Terro-rism information extraction from online reports[J]. Journal of Computer Information Systems, 2015, 55(3): 20-28.
[5] ATKINSON M, PISKORSKI J, TANEV H, et al. Automa-ted event extraction in the domain of border security[C]//Proceedings of the 1st International Conference on User Centric Media,Venice, Dec 9-11, 2009. Berlin, Heidelberg:Springer, 2009: 321-326.
[6] TANEV H, PISKORSKI J, ATKINSON M. Real-time news event extraction for global crisis monitoring[C]//LNCS 5039: Proceedings of the 13th International Conference on Applications of Natural Language to Information Systems, London, Jun 24-27, 2008. Berlin, Heidelberg: Springer, 2008: 207-218.
[7] ATKINSON M, DU M, PISKORSKI J, et al. Techniques for multilingual security-related event extraction from on-line news[M]//PRZEPIóRKOWSKI A, PIASECKI M, JAS-SEM K, eds. Computational Linguistics. Berlin, Heidel-berg: Springer, 2013: 163-186.
[8] PISKORSKI J, TANEV H, OEZDENWENNERBERG P. Ex-tracting violent events from on-line news for ontology po-pulation[C]//LNCS 4439: Proceedings of the 10th Interna-tional Conference on Business Information Systems, Poznan, Apr 25-27, 2007. Berlin, Heidelberg: Springer, 2007: 287-300.
[9] NUIJ W, MILEA V, HOGENBOOM F, et al. An automated framework for incorporating news into stock trading stra-tegies[J]. IEEE Transactions on Knowledge and Data Engi-neering, 2013, 26(4): 823-835.
[10] CAPET P, DELAVALLADE T, NAKAMURA T, et al. A risk assessment system with automatic extraction of event types[C]//Proceedings of the 5th IFIP International Conference on Intelligent Information Processing, Beijing, Oct 19-22, 2008. Berlin, Heidelberg: Springer, 2009: 220-229.
[11] ABRAHAMS A S. Developing and executing electronic commerce applications with occurrences[D]. Cambridge: Uni-versity of Cambridge, 2002.
[12] VANEGAS J A, MATOS S, GONZáLEZ F, et al. An over-view of biomolecular event extraction from scientific docu-ments[J]. Computational and Mathematical Methods in Medicine, 2015.
[13] DODDINGTON G R, MITCHELL A, PRZYBOCKI M A, et al. The automatic content extraction (ACE) program- tasks, data, and evaluation[C]//Proceedings of the 4th Inter-national Conference on Language Resources and Eval-uation, Lisbon, May 26-28, 2004. Paris: European Langua-ge Resources Association, 2004: 837-840.
[14] AHN D. The stages of event extraction[C]//Proceedings of the 2006 Workshop on Annotating and Reasoning about Time and Events, Sydney, Jul 23, 2006. Stroudsburg: ACL, 2006: 1-8.
[15] LAMPLE G, BALLESTEROS M, SUBRAMANIAN S, et al. Neural architectures for named entity recognition[J]. arXiv:1603.01360, 2016.
[16] YIH W, HE X D, MEEK C. Semantic parsing for single-rela-tion question answering[C]//Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, Baltimore, Jun 22-27, 2014. Stroudsburg: ACL, 2014: 643-648.
[17] SHEN Y L, HE X D, GAO J F, et al. Learning semantic representations using convolutional neural networks for web search[C]//Proceedings of the 23rd International World Wide Web Conference, Seoul, Apr 7-11, 2014. New York: ACM, 2014: 373-374.
[18] ZHANG Y, WALLACE B. A sensitivity analysis of (and practitioners?? guide to) convolutional neural networks for sentence classification[J]. arXiv:1510.03820, 2015.
[19] KALCHBRENNER N, GREFENSTETTE E, BLUNSOM P. A convolutional neural network for modelling sentences[J]. arXiv:1404.2188, 2014.
[20] COLLOBERT R, WESTON J, BOTTOU L, et al. Natural lan-guage processing (almost) from scratch[J]. Journal of Ma-chine Learning Research, 2011, 12: 2493-2537.
[21] ZENG D J, LIU K, LAI S W, et al. Relation classification via convolutional deep neural network[C]//Proceedings of the 25th International Conference on Computational Lingui-stics, Dublin, Aug 23-29, 2014. Stroudsburg: ACL, 2014: 2335-2344.
[22] NGUYEN T H, GRISHMAN R. Relation extraction: pers-pective from convolutional neural networks[C]//Proceedings of the 1st Workshop on Vector Space Modeling for Natural Language Processing, Denver, Jun 5, 2015. Stroudsburg: ACL, 2015: 39-48.
[23] MIKOLOV T, CHEN K, CORRADO G, et al. Efficient esti-mation of word representations in vector space[J]. arXiv:1301.3781, 2013.
[24] MIKOLOV T, SUTSKEVER I, CHEN K, et al. Distributed representations of words and phrases and their composi-tionality[C]//Proceedings of the 27th Annual Conference on Neural Information Processing Systems 2013, Lake Tahoe, Dec 5-8, 2013. Red Hook: Curran Associates, 2013: 3111-3119.
[25] CHEN Y B, XU L H, LIU K, et al. Event extraction via dynamic multi-pooling convolutional neural networks[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing, Beijing, Jul 26-31, 2015. Stroudsburg: ACL, 2015: 167-176.
[26] ZHANG Z K, XU W R, CHEN Q Q. Joint event extraction based on skip-window convolutional neural networks[C]//LNCS 10102: Proceedings of the 5th CCF Conference on Natural Language Processing and Chinese Computing, and 24th International Conference on Computer Processing of Oriental Languages, Kunming, Dec 2-6, 2016. Cham: Sprin-ger, 2016: 324-334.
[27] BUREL G, SAIF H, FERNANDEZ M, et al. On semantics and deep learning for event detection in crisis situations[C]//Proceedings of the 2017 Workshop on Semantic Deep Lear-ning, Portoroz, May 29, 2017. UK: The Open University, 2017: 1-12.
[28] LI L, LIU Y, QIN M. Extracting biomedical events with parallel multi-pooling convolutional neural networks[J]. IEEE/ACM Transactions on Computational Biology and Bioinfor-matics, 2018, 17(2): 599-607.
[29] KODELJA D, BESAN?ON R, FERRET O. Exploiting a more global context for event detection through bootstrap-ping[C]//LNCS 11437: Proceedings of the 41st European Conference on IR Research, Cologne, Apr 14-18, 2019. Cham: Springer, 2019: 763-770.
[30] MIKOLOV T, KARAFIáT M, BURGET L, et al. Recur-rent neural network based language model[C]//Proceedings of the 11th Annual Conference of the International Speech Communication Association, Makuhari, Sep 26-30, 2010: 1045-1048.
[31] FENG X, QIN B, LIU T. A language-independent neural network for event detection[J]. Science China Information Sciences, 2018, 61(9): 1-12.
[32] NGUYEN T H, CHO K, GRISHMAN R. Joint event extrac-tion via recurrent neural networks[C]//Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua-ge Technologies, San Diego, Jun 12-17, 2016. Stroudsburg: ACL, 2016: 300-309.
[33] CHO K, VAN MERRI?NBOER B, GULCEHRE C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[J]. arXiv:1406.1078, 2014.
[34] SHA L, QIAN F, CHANG B B, et al. Jointly extracting event triggers and arguments by dependency-bridge RNN and tensor-based argument interaction[C]//Proceedings of the 32nd AAAI Conference on Artificial Intelligence, the 30th Inno-vative Applications of Artificial Intelligence, and the 8th AAAI Symposium on Educational Advances in Artificial Intelligence, New Orleans, Feb 2-7, 2018. Menlo Park: AAAI, 2018: 5916-5923.
[35] TAI K S, SOCHER R, MANNING C D. Improved seman-tic representations from tree-structured long short-term me-mory networks[J]. arXiv:1503.00075, 2015.
[36] ZHANG W B, DING X, LIU T. Learning target-dependent sentence representations for Chinese event detection[C]//Proceedings of the 24th China Conference on Information Retrieval, Guilin, Sep 27-29, 2018. Cham: Springer, 2018: 251-262.
[37] LI D Y, HUANG L F, JI H, et al. Biomedical event extrac-tion based on knowledge-driven tree-LSTM[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Lan-guage Technologies, Minneapolis, Jun 2-7, 2019. Strouds-burg: ACL, 2019: 1421-1430.
[38] LIU X, LUO Z, HUANG H. Jointly multiple events extrac-tion via attention-based graph information aggregation[J]. arXiv:1809.09078, 2018.
[39] RAO S, MARCU D, KNIGHT K, et al. Biomedical event ex-traction using abstract meaning representation[C]//Procee-dings of the Biomedical Natural Language Processing Workshop 2017, Aug 4, 2017. Stroudsburg: ACL, 2017: 126-135.
[40] SCARSELLI F, GORI M, TSOI A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2008, 20(1): 61-80.
[41] WU Z, PAN S, CHEN F, et al. A comprehensive survey on graph neural networks[J]. IEEE Transactions on Neural Net-works and Learning Systems, 2020, 32(1): 4-24.
[42] ZHOU J, CUI G, HU S, et al. Graph neural networks: a re-view of methods and applications[J]. AI Open, 2020, 1: 57-81.
[43] PETERS M E, NEUMANN M, IYYER M, et al. Deep con-textualized word representations[J]. arXiv:1802.05365, 2018.
[44] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language unders-tanding[J]. arXiv:1810.04805, 2018.
[45] LI X Y, LI F Y, PAN L, et al. DuEE: a large-scale dataset for Chinese event extraction in real-world scenarios[C]//LNCS 12431: Proceedings of the 9th CCF International Conferen-ce on Natural Language Processing and Chinese Computing, Zhengzhou, Oct 14-18, 2020. Cham: Springer, 2020: 534-545.
[46] 王炳乾, 宿绍勋, 梁天新. 基于 BERT 的多层标签指针网络事件抽取模型——2020 语言与智能技术竞赛事件抽取任务系统报告[J]. 中文信息学报, 2021, 35(7): 81-88.
WANG B Q, SU S X, LIANG T X. BERT based multi-layer label pointer network for event extraction—2020 language and intelligence technology competition event extraction task system report[J]. Journal of Chinese Information Pro-cessing, 2021, 35(7): 81-88.
[47] DU X, CARDIE C. Event extraction by answering (almost) natural questions[J]. arXiv:2004.13625, 2020.
[48] LIU J, CHEN Y F, XU J N. Machine reading comprehension as data augmentation: a case study on implicit event argu-ment extraction[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Punta Cana, Nov 7-11, 2021. Stroudsburg: ACL, 2021: 2716-2725.
[49] LI S, JI H, HAN J. Document-level event argument extrac-tion by conditional generation[J]. arXiv:2104.05919, 2021.
[50] LEWIS M, LIU Y, GOYAL N, et al. BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension[J]. arXiv:1910.13461, 2019.
[51] RAFFEL C, SHAZEER N, ROBERTS A, et al. Exploring the limits of transfer learning with a unified text-to-text trans-former[J]. Journal of Machine Learning Research, 2020, 21: 1-67.
[52] HSU I H, HUANG K H, BOSCHEE E, et al. DEGREE: a data-efficient generation-based event extraction model[C]//Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Seattle, Jul 10-15, 2022.Stroudsburg: ACL, 2022: 1890-1908.
[53] LIU X, HUANG H, SHI G, et al. Dynamic prefix-tuning for generative template-based event extraction[J]. arXiv:2205.06166, 2022.
[54] MA Y, WANG Z, CAO Y, et al. Prompt for extraction? PAIE: prompting argument interaction for event argument extrac-tion[J]. arXiv:2202.12109, 2022.
[55] LIU S L, LI Y, ZHANG F, et al. Event detection without triggers[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computa-tional Linguistics: Human Language Technologies, Minnea-polis, Jun 2-7, 2019. Stroudsburg: ACL, 2019: 735-744.
[56] ZHENG S, CAO W, XU W, et al. Doc2EDAG: an end-to-end document-level framework for Chinese financial event extraction[J]. arXiv:1904.07535, 2019.
[57] 王雷, 李瑞轩, 李玉华, 等. 文档级无触发词事件抽取联合模型[J]. 计算机科学与探索, 2021, 15(12): 2327-2334.
WANG L, LI R X, LI Y H, et al. A joint model for document-level event extraction without triggers[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(12): 2327-2334.
[58] XU R, LIU T, LI L, et al. Document-level event extraction via heterogeneous graph-based interaction model with a tracker[J]. arXiv:2105.14924, 2021.
[59] DENG S M, ZHANG N Y, KANG J J, et al. Meta-learning with dynamic-memory-based prototypical network for few-shot event detection[C]//Proceedings of the 13th Interna-tional Conference on Web Search and Data Mining, Hous-ton, Feb 3-7, 2020. New York: ACM, 2020: 151-159.
[60] LAI V D, DERNONCOURT F, NGUYEN T H. Extensively matching for few-shot learning event detection[J]. arXiv:2006.10093, 2020.
[61] LIU S L, CHEN Y B, HE S Z, et al. Leveraging FrameNet to improve automatic event detection[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Berlin, Aug 7-12, 2016. New York: ACM, 2016: 2134-2143.
[62] CHEN Y B, LIU S L, ZHANG X, et al. Automatically labe-led data generation for large scale event extraction[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, Vancouver, Aug 4, 2017. Stroudsburg: ACL, 2017: 409-419.
[63] TONG M H, XU B, WANG S, et al. Improving event detec-tion via open-domain event trigger knowledge[C]//Procee-dings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2020: 5887-5897.
[64] LINGUISTIC D C. Rich ERE annotation guidelines over-view[R]. 2015.
[65] KIM J D, OHTA T, TATEISI Y, et al. GENIA corpus—a semantically annotated corpus for bio-textmining[J]. Bioin-formatics, 2003, 19: 180-182.
[66] NéDELLEC C, BOSSY R, KIM J D, et al. Overview of BioNLP shared task 2013[C]//Proceedings of the BioNLP Shared Task 2013 Workshop, Sofia, Aug 9, 2013. Stroud-sburg: ACL, 2013: 1-7.