计算机科学 ›› 2020, Vol. 47 ›› Issue (12): 233-238.doi: 10.11896/jsjkx.191100031

• 人工智能 • 上一篇    下一篇

基于篇章信息和Bi-GRU的中文事件检测

朱培培, 王中卿, 李寿山, 王红玲   

  1. 苏州大学计算机科学与技术学院 江苏 苏州 215006
  • 收稿日期:2019-11-05 修回日期:2019-12-30 发布日期:2020-12-17
  • 通讯作者: 王中卿(wangzq@suda.edu.cn)
  • 作者简介:ppzhu@stu.suda.edu.cn
  • 基金资助:
    国家自然科学基金青年科学基金(6180613761702518);江苏省高校自然科学研究基金(18KJB520043)

Chinese Event Detection Based on Document Information and Bi-GRU

ZHU Pei-pei, WANG Zhong-qing, LI Shou-shan, WANG Hong-ling   

  1. School of Computer Science and Technology Soochow University Suzhou Jiangsu 215006,China
  • Received:2019-11-05 Revised:2019-12-30 Published:2020-12-17
  • About author:ZHU Pei-pei,born in 1995postgra-duateis a member of China Computer Federation.Her main research interests include natu-ral language processing and so on.
    WANG Zhong-qing,born in 1987Ph.D.His main research interests include natural language processing and so on.
  • Supported by:
    Young Scientists Fund of the National Natural Science Foundation of China(61806137,61702518) and Natural Science Foundation of the Higher Education Institutions of Jiangsu Province,China (18KJB520043).

摘要: 事件抽取是信息抽取中一个重要的研究方向其中事件检测是事件抽取的关键.目前中文神经网络事件检测方法均是基于句子的方法这种方法获得的局部上下文的信息不足以解决事件触发词的歧义性.针对这个问题文中探索了篇章信息的作用.首先以双向门控循环单元网络(Bidirectional Gated Recurrent UnitsBi-GRU)模型为基线定义3个窗口来学习句子特征;然后将句子表示进行拼接利用双向门控循环单元网络学习句子的上下文特征;最后将句子表示和上下文表示进行融合以丰富句子的语义信息并减少候选触发词语义模糊现象通过Softmax函数进行事件触发词的分类.在ACE2005数据集上的实验结果表明句子的上下文特征能够有效提升中文事件检测方法的性能该中文事件检测方法的F1值比当前最好的模型高1.5%.

关键词: ACE2005, 篇章信息, 事件抽取, 事件检测, 双向门控循环单元网络

Abstract: Event extraction is an important research task in information extraction and event detection is the key to event extraction.Existing Chinese neural network event detection methods are sentence-based and the local context information obtained by this method is not enough to resolve the event triggers semantic ambiguity.In order to solve this problemthis paper studies document information effects.Firstlybased on the bidirectional gated recurrent units network (Bi-GRU)this paper defines three windows to learn sentence features.Thenthe sentence-level representation is concatenated and the document features are learned by using the bidirectional gated recurrentunits network.Finallyto enrich the semantic information of sentences and reduce the event-trigger sematic event triggers ambiguityit merges the sentence-level representation and the document-level representation and then classifies eventtriggers through the Softmax function.Experimental results on the ACE2005 dataset show that the sentences-context representation can improve the Chinese event detection performance and this event detection method outperforms state-of-the-art results by 1.5% on F1.

Key words: ACE2005, Bidirectional gated recurrent units, Document information, Event detection, Event extraction

中图分类号: 

  • TP391
[1] CHEN Z,JI H.Language specific issue and feature exploration in chinese event extraction[C]//Proceedings of 47th ACL.Boulder,Colorado:Association for Computational Linguistics,2009:209-212.
[2] LI P F,ZHOU G D.Employing morphological structures andsememes for chinese event extration[C]//Proceedings of COLING 2012.Mumbai,India,2012:1619-1634.
[3] WANG W.Chinese news event 5w1h semantic elements extraction for event ontology population[C]//Proceedings of the WWW 2012.Lyon,France,2012:197-202.
[4] JI H.Cross-lingual Predicate Cluster Acquisition to improve Bilingual Event Extraction by Inductive Learning[C]//Proceedings of the Workshop on UMSLLS 2009.Boulder,Colorado:Association for Computational Linguistics,2009:27-35.
[5] LI P F,ZHU Q M,DIAO H J,et al.Joint modeling of triggeridentification and event type determination in chinese event extraction[C]//Proceedings of COLING 2012.Mumbai,India,2012:1635-1651.
[6] LI Q,JI H,HUANG L.Joint event extraction via structured prediction with global features[C]//Proceedings of the 51st ACL.Sofia,Bulgaria:Association for Computational Linguistics,2013:789-797.
[7] NGUYEN T H,GRISHMAN R.Event detection and domain adaptation withconvolutional neural networks[C]//Proceedings of 53rd ACL.Beijing,China:Association for Computational Linguistics,2015:365-371.
[8] CHEN Y B,XU L H,LIU K,et al.Event extraction via dynamic multi-poolingconvolutional neural networks[C]//Proceedings of 53rd ACL.Beijing,China:Association for Computational Linguistics,2015:167-176.
[9] NGUYEN T H,CHO K,GRISHMAN R.Joint event extraction via recurrent neural networks[C]//Proceedings of NAACL-HLT 2016.San Diego,California,USA:Association for Computational Linguistics,2016:300-309.
[10] FENG X C,HUANG L F,TANG D Y,et al.A language independent neural network for event detection[C]//Proceedings of 54th ACL.Berlin,Germany:Association for Computational Linguistics,2016:66-71.
[11] LIU S L,CHEN Y B,LIUK,et al.Exploiting argument information to improve event detection via supervised attention mechanisms[C]//Proceedings of 55th ACL.Vancouver,Canada:Association for Computational Linguistics,2017:1789-1798.
[12] LIU S L,CHEN Y B,HE S Z,et al.Leveraging framenet to improve automatic event detection[C]//Proceedings of 54th ACL.Berlin,Germany:Association for Computational Linguistics,2016:2134-2143.
[13] LIN H Y,LU Y J,HAN X P,et al.Nugget proposal networks forchinese event detection[C]//Proceedings of 56th ACL.Melbourne,Australia:Association for Computational Linguistics,2018:1565-1574.
[14] ZENG Y,YANG H H,FENG Y S,et al.A convolution bilstm neural network model for chinese event extraction[C]//Proceedings of NLPCC-ICCPOL.2016:275-287.
[15] ZENG Y,FENG Y S,MA R,et al.Scale up event extractionlearning via automatic training data generation[C]//Proceedings of 32nd AAAI Conference on Artificial Intelligence.New Orleans,Louisiana,USA:AAAI Press,2018:6045-6052.
[16] JI H,GRISHMAN R.Refining event extraction through cross-document inference[C]//Proceedings of 46th ACL.Columbus,Ohio,USA:Association for Computational Linguistics,2008:254-262.
[17] LIAO S S,GRISHMAN R.Using document level cross-event inference to improve event extraction[C]//Proceedings of 48th ACL.Uppsala,Sweden:Association for Computational Linguistics,2010:789-797.
[18] HONG Y,ZHANG J F,MA B,et al.Using cross-entity inference to improve event extraction[C]//Proceedings of 49th ACL.Portland,Oregon,USA:Association for Computational Linguistics,2011:1127-1136.
[19] ZHAO Y,JIN X L,WANG Y Z,et al.Document embedding enhanced event detection with hierarchical and supervised attention[C]//Proceedings of 56th ACL.Melbourne,Australia:Association for Computational Linguistics,2018:414-419.
[20] LI P F,ZHOU G D,ZHU Q M,et al.Employing compositional semantics and discourse consistency inchinese event extraction[C]//Proceedings of EMNLP 2012.Jeju Island,Korea:Association for Computational Linguistics,2012:1006-1016.
[21] CHEN C,NG V.Joint modeling ofchinese event extraction with rich linguistic features[C]//Proceedings of COLING 2012.Mumbai,India,2012:529-544.
[1] 孙奇, 吉根林, 张杰.
基于非局部注意力生成对抗网络的视频异常事件检测方法
Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection
计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[2] 缪峰, 王萍, 李太勇.
基于事件动作方向的隐式因果关系抽取方法
Implicit Causality Extraction Method Based on Event Action Direction
计算机科学, 2022, 49(3): 276-280. https://doi.org/10.11896/jsjkx.211100249
[3] 刘文洋, 郭延哺, 李维华.
识别关键蛋白质的混合深度学习模型
Identifying Essential Proteins by Hybrid Deep Learning Model
计算机科学, 2021, 48(8): 240-245. https://doi.org/10.11896/jsjkx.200700076
[4] 侯春萍, 赵春月, 王致芃.
基于自反馈最优子类挖掘的视频异常检测算法
Video Abnormal Event Detection Algorithm Based on Self-feedback Optimal Subclass Mining
计算机科学, 2021, 48(7): 199-205. https://doi.org/10.11896/jsjkx.200800146
[5] 卿来云, 张建功, 苗军.
在线异常事件检测的时序建模
Temporal Modeling for Online Anomaly Detection
计算机科学, 2021, 48(7): 206-212. https://doi.org/10.11896/jsjkx.200900093
[6] 丁玲, 向阳.
基于分层次多粒度语义融合的中文事件检测
Chinese Event Detection with Hierarchical and Multi-granularity Semantic Fusion
计算机科学, 2021, 48(5): 202-208. https://doi.org/10.11896/jsjkx.200800038
[7] 吴凡, 朱培培, 王中卿, 李培峰, 朱巧明.
基于字词联合表示的中文事件检测方法
Chinese Event Detection with Joint Representation of Characters and Words
计算机科学, 2021, 48(4): 249-253. https://doi.org/10.11896/jsjkx.200300156
[8] 余杰, 纪斌, 刘磊, 李莎莎, 马俊, 刘慧君.
面向中文医疗事件的联合抽取方法
Joint Extraction Method for Chinese Medical Events
计算机科学, 2021, 48(11): 287-293. https://doi.org/10.11896/jsjkx.201200016
[9] 姚兰, 赵永恒, 施雨晴, 于明鹤.
一种基于视频分析的高速公路交通异常事件检测算法
Highway Abnormal Event Detection Algorithm Based on Video Analysis
计算机科学, 2020, 47(8): 208-212. https://doi.org/10.11896/jsjkx.191000165
[10] 富坤, 仇倩, 赵晓梦, 高金辉.
基于节点演化分阶段优化的事件检测方法
Event Detection Method Based on Node Evolution Staged Optimization
计算机科学, 2020, 47(5): 96-102. https://doi.org/10.11896/jsjkx.190400072
[11] 高李政, 周刚, 黄永忠, 罗军勇, 王树伟.
基于Zipf's共生矩阵分解的开放域事件向量计算方法
Open Domain Event Vector Algorithm Based on Zipf's Co-occurrence Matrix Factorization
计算机科学, 2020, 47(10): 207-214. https://doi.org/10.11896/jsjkx.191200183
[12] 高利剑,毛启容.
环境辅助的多任务混合声音事件检测方法
Environment-assisted Multi-task Learning for Polyphonic Acoustic Event Detection
计算机科学, 2020, 47(1): 159-164. https://doi.org/10.11896/jsjkx.190200365
[13] 高李政, 周刚, 罗军勇, 兰明敬.
元事件抽取研究综述
Survey on Meta-event Extraction
计算机科学, 2019, 46(8): 9-15. https://doi.org/10.11896/j.issn.1002-137X.2019.08.002
[14] 李志国, 钟将, 钟璐蔓.
复杂事件管理的多元时序数据处理技术研究
Study on Processing Technology for Complex Event Management Based on Multivariate Time Series Data
计算机科学, 2019, 46(6): 55-63. https://doi.org/10.11896/j.issn.1002-137X.2019.06.007
[15] 张爱英,倪崇嘉.
基于音频事件检测和分类的音频监控系统背景模型自适应方法研究
Research on Background Model Adaptation for Acoustic Event Detection and Classification Based on Acoustic Surveillance System
计算机科学, 2016, 43(9): 310-314. https://doi.org/10.11896/j.issn.1002-137X.2016.09.062
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!