文档级无触发词事件抽取联合模型

doi:10.3778/j.issn.1673-9418.2008097

摘要/Abstract

摘要：

当前广为研究的在单个句子范围内的事件抽取方法，难以扩展到从分布在一篇文章里的多个句子中抽取同一事件的完整论元。对此，提出了一种基于深度学习的文档级事件抽取联合模型。首先，利用基于多头自注意力机制的实体识别模块逐句识别文档中的实体并输出其类型。然后，通过定义不同论元角色对事件类型的重要度训练事件类型检测模块，实现在无触发词条件下定位事件表述中心句并判断事件类型。最后，事件论元抽取模块通过在实体语义向量中嵌入实体的类型信息和实体到事件中心句的距离信息，并输入Transformer网络与上下文交换信息，实现在文档范围内抽取全部事件论元。通过对上述三个子模块进行联合训练，进一步实现了端到端的事件抽取，避免了管道式方法的误差传递。在公开数据集上的实验结果表明：在单事件条件下，该模型取得了86.3%的[F1]值，优于当前最佳的文档级事件抽取方法，并且具有优秀的模型训练速度。

关键词: 文档级事件抽取, 无触发词, 联合模型, 实体识别, 事件检测

Abstract:

The widely researched sentence-level event extraction methods struggle to extract all arguments of the same event from a whole document. To solve this problem, this paper proposes a joint model for document-level event extraction based on deep learning. Firstly, an entity recognition module based on multi-head self-attention mechanism is used to identify entities and their types sentence by sentence. Then, an event type detection module trained by defining the importance of different argument roles, is used to locate the event mention sentence and predict the event type without the help of event triggers. Finally, an event argument extraction module embeds every entity??s semantic vector with its type information and its distance to the event mention sentence before feeding into a context-aware Transformer, in order to extract arguments within the document scope. In addition, by training the three modules mentioned above jointly, this paper realizes an end-to-end event extraction model and avoids error propagation problems in traditional pipeline models. The experimental results on a public dataset shows that, when each document contains only one event, the proposed model achieves a 86.3% [F1-score], which outperforms state-of-the-art methods, and the training process completes rather quickly.

Key words: document-level event extraction, triggers free, joint model, entity recognition, event detection

王雷, 李瑞轩, 李玉华, 辜希武, 杨琪. 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 Lei, LI Ruixuan, LI Yuhua, GU Xiwu, YANG Qi. Joint Model for Document-Level Event Extraction Without Triggers[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(12): 2327-2334.

参考文献

[1] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st Annual Conference on Neural Information Processing Systems, Long Beach, Dec 4-9, 2017. Red Hook: Curran Associates, 2017: 5998-6008.
[2] RILOFF E. Automatically constructing a dictionary for inf-ormation extraction tasks[C]//Proceedings of the 11th Nati-onal Conference on Artificial Intelligence, Washington, Jul 11-15, 1993. Menlo Park: AAAI, 1993: 811-816.
[3] RILOFF E, SHOEN J. Automatically acquiring conceptual patterns without an annotated corpus[C]//Proceedings of the 3rd Workshop on Very Large Corpora, Cambridge, Jun 30, 1995. Stroudsburg: ACL, 1995: 148-161.
[4] JIANG J F. A research about the pattern acquisition for free text IE[D]. Beijing: Graduate University of Chinese Aca-demy of Sciences (Institute of Computing Technology), 2004.
姜吉发. 自由文本的信息抽取模式获取的研究[D]. 北京：中国科学院研究生院(计算技术研究所), 2004.
[5] CHIEU H L, NG H T. A maximum entropy approach to in-formation extraction from semi-structured and free text[C]//Proceedings of the 18th National Conference on Artificial Intelligence and 14th Conference on Innovative Applica-tions of Artificial Intelligence, Edmonton, Jul 28-Aug 1, 2002. Menlo Park: AAAI, 2002: 786-791.
[6] 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.
[7] ZHAO Y Y, QIN B, CHE W X, et al. Research on Chinese event extraction[J]. Journal of Chinese Information Proces-sing, 2008, 22(1): 3-8.
赵妍妍, 秦兵, 车万翔, 等. 中文事件抽取技术研究[J]. 中文信息学报, 2008, 22(1): 3-8.
[8] 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.
[9] YANG S, FENG D W, QIAO L B, et al. Exploring pre-trained language models for event extraction and generation[C]//Proceedings of the 57th Conference of the Association for Computational Linguistics, Florence, Jul 28-Aug 2, 2019. Stroudsburg: ACL, 2019: 5284-5294.
[10] NGUYEN T H, CHO K, GRISHMAN R. Joint event extraction via recurrent neural networks[C]//Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, San Diego, Jun 12-17, 2016. Stroudsburg: ACL, 2016: 300-309.
[11] SHA L, QIAN F, CHANG B B, et al. Jointly extracting ev-ent triggers and arguments by dependency-bridge RNN and tensor-based argument interaction[C]//Proceedings of the 32nd AAAI Conference on Artificial Intelligence, the 30th Innovative Applications of Artificial Intelligence, and the 8th AAAI Symposium on Educational Advances in Artifi-cial Intelligence, New Orleans, Feb 2-7, 2018. Menlo Park: AAAI, 2018: 5916-5923.
[12] CHEN Y B, LIU S L, ZHANG X, et al. Automatically labeled data generation for large scale event extraction[C]// Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, Vancouver, Jul 30-Aug 4,2017. Stroudsburg: ACL, 2017: 409-419.
[13] MINTZ M, BILLS S, SNOW R, et al. Distant supervision for relation extraction without labeled data[C]//Proceedings of the 47th Annual Meeting of the Association for Compu-tational Linguistics, and the 4th International Joint Con-ference on Natural Language Processing of the AFNLP, Singapore, Aug 2-7, 2009. Stroudsburg: ACL, 2009: 1003-1011.
[14] ZENG Y, FENG Y S, MA R, et al. Scale up event extraction learning via automatic training data generation[C]//Procee-dings of the 32nd AAAI Conference on Artificial Intelli-gence, the 30th Innovative Applications of Artificial Intelli-gence, and the 8th AAAI Symposium on Educational Ad-vances in Artificial Intelligence, New Orleans, Feb 2-7, 2018. Menlo Park: AAAI, 2018: 6045-6052.
[15] YANG H, CHEN Y B, LIU K, et al. DCFEE: a document-level Chinese financial event extraction system based on automatically labeled training data[C]//Proceedings of ACL 2018, Melbourne, Jul 15-20, 2018. Stroudsburg: ACL, 2018: 50-55.
[16] ZHONG W F, YANG H, CHEN Y B, et al. Document-level event extraction based on joint labeling and global reasoning[J]. Journal of Chinese Information Processing, 2019, 33(9): 88-95.
仲伟峰, 杨航, 陈玉博, 等. 基于联合标注和全局推理的篇章级事件抽取[J]. 中文信息学报, 2019, 33(9): 88-95.
[17] 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.
[18] BOJANOWSKI P, GRAVE E, JOULIN A, et al. Enriching word vectors with subword information[J]. Transactions of the Association for Computational Linguistics, 2017, 5: 135-146.
[19] ZHANG Y, YANG J. Chinese NER using lattice LSTM[J]. arXiv:1805.02023, 2018.
[20] YAN H, DENG B C, LI X N, et al. TENER: adapting transformer encoder for name entity recognition[J]. arXiv:1911.04474, 2019.