事件同指消解是自然语言处理中一个具有挑战性的任务,它在事件抽取、问答系统和阅读理解中具有重要作用。针对事件的语义信息主要由触发词和论元表示这一个特点,该文将事件进行结构化表示并输入一个基于门控和注意力机制的模型GAN-SR(gated attention network with structured representation),在文档内进行中文事件同指消解。首先,该模型采用语义角色标注和依存句法分析技术对事件句进行浅层语义分析,抽取事件句信息并表示为一个事件五元组。其次,将各种事件信息输入GRU进行编码,然后使用多头注意力机制挖掘事件句和事件对之间的重要特征。在ACE2005中文语料库上的实验表明,GAN-SR的性能优于目前性能最好的基准系统。
Abstract
Event coreference resolution is a challenging task in natural language processing. Since the semantics of an event is mainly represented by its trigger and arguments, this paper introduces the structured representation of events to a neural network model GAN-SR (gated attention network with structured representation) based on the gated and attention mechanism for Chinese event coreference resolution in document level. Firstly, we apply the semantic role labeling and dependency parsing to analyze the shallow semantics of events, and then uses a quintile to represent their structures. Secondly, we use GRU to encode various kinds of event information and then apply the multi-head attention mechanism to mine the important features between the events and event pairs. The experimental results on the ACE2005 Chinese corpus show that GAN-SR outperforms the state-of-the-art baselines.
关键词
中文事件同指消解 /
结构化表示 /
注意力机制
{{custom_keyword}} /
Key words
Chinese event coreference resolution /
structured representation /
attention mechanism
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Mohammed M Fouad, Marwa A Atyah. Efficient topic detection system for online arabic news[J]. International Journal of Computer Applications, 2018, 180(12): 7-12.
[2] 程良, 郜洪奎, 王红斌. 基于依存树与规则相结合的汉泰新闻事件要素抽取方法[J]. 软件导刊, 2018, 17(07): 53-60,67.
[3] Swabha Swayamdipta, Ankur P Parikh, Tom Kwiat-kowski. Multi-mention learning for reading comprehension with neural cascades[C]//Proceedings of the ICLR, 2018.
[4] Sebastian Krause, Feiyu Xu, Hans Uszkoreit, et al. Event linking with sentential features from convolutional neural networks[C]//Proceedings of the Signal Conference on Computational Natural Language Learning, 2016: 239-249.
[5] Jie Fang, Peifeng Li, Guodong Zhou. Employing multiple decomposable attention networks to resolve event coreference[C]//Proceedings of the CCF International Conference on Natural Language Processing and Chinese Computing, 2018: 246-256.
[6] David Ahn. The stages of event extraction[C]//Proceedings of the Workshop of the ACL on Annotating and Reasoning about Time and Events, 2006: 1-8.
[7] Haghighi Aria,Klein Dan. Simple coreference resolution with rich syntactic and semantic features[C]//Proceedings of the EMNLP, 2009: 1152-1161.
[8] Agata Cybulska, Piek Vossen. Translating granularity of event slots into features for event coreference resolution[C]//Proceedings of the 3rd Workshop on EVENTS: Definition, Detection, Coreference, and Representation, 2015: 1-10.
[9] Zheng Chen, Heng Ji, Robert Haralick. A pairwise event coreference model, feature impact and evaluation for event coreference resolution[C]//Proceedings of the Workshop on Events in Emerging Text Types, 2009: 17-22.
[10] Chen Chen, Vincent Ng. Sino Coreferencer: An end-to-end Chinese event coreference resolver[C]//Proceedings of the 9th Language Resources and Evaluation Conference, 2004: 4532-4538.
[11] Thien Huu Nguyen, Adam Meyers, Ralph Grishman. New York university 2016 system for KBP event nugget: A deep learning approach[C]//Proceedings of the Text Analysis Conference, 2016: 45-52.
[12] Cosmin Adrian Bejan, SandaHarabagiu. Unsupervised event coreference resolution[J]. Computational Linguistics, 2014, 40(2): 1412-1422.
[13] Zhengzhong Liu, Jun Araki, Eduard Hovy, et al. Supervised within document event coreference using information propagation[C]//Proceedings of the 9th Language Resources and Evaluation Conference, 2014: 4539-4544.
[14] 滕佳月,李培峰,朱巧明.基于全局优化的中文事件同指消解方法[J]. 北京大学学报(自然科学版), 2016, 52(1): 97-103
[15] Haoruo Peng, Yangqi Song, Dan Roth. Event detection and coreference with minimal supervision[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2016: 392-402.
[16] Ashish Vaswani, Noam Shazeer, NikiParmar. Attention is all you need[C]//Proceedings of the NIPS 2017, 2017: 61-75.
[17] Wanxiang Che, Zhenghua Li, Ting Liu. LTP: A Chinese language technology platform[C]//Proceedings of the Coling, 2010: 13-16.
[18] 滕佳月. 中文事件同指消解方法研究[D].苏州: 苏州大学硕士学位论文, 2016.
[19] Marc Vilain, John Burger, John Aberdeen, et al. A model-theoretic coreference scoring scheme[C]//Proceedings of the MUC-6, 1995:45-52.
[20] Amit Bagga, Breck Baldwin. Algorithms for scoring coreference chains[C]//Proceedings of the LREC 1998, 1998: 563-566.
[21] Marta Recasen, Eduard Hovy. Blanc: Implementing the rand index for coreference evaluation[J]. Natural Language Engineering, 2011: 485-510.
[22] Xiaoqiang Luo. On coreference resolution performance metrics[C]//Proceedings of the EMNLP 2005, 2005: 25-32.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61836007,61772354,61773276)
{{custom_fund}}
PDF(2008 KB)
