基于门控记忆网络的汉语篇章主次关系识别方法  

GMN-based Nuclearity Recognition in Chinese Discourse

作　　者：王体爽 李培峰[1] 朱巧明[1] WANG Tishuang;LI Peifeng;ZHU Qiaoming(School of Computer Science and Technology,Soochow University,Suzhou,Jiangsu 215006,China)

机构地区：[1]苏州大学计算机科学与技术学院,江苏苏州215006

出　　处：《中文信息学报》2019年第5期39-46,共8页Journal of Chinese Information Processing

基　　金：国家自然科学基金(61836007;61772354;61773276)

摘　　要：篇章分析是自然语言理解的基础。作为篇章分析的重要任务之一,汉语主次关系识别还处于探索阶段。该文提出了一种基于门控记忆网络(GMN)的汉语篇章主次关系识别方法。该方法首先使用Bi-LSTM和CNN分别获取每个篇章单元的全局信息和局部信息。然后,融合两部分篇章单元信息并从中计算得到一个门控单元。最后,使用这个门控单元捕获各个篇章单元相对于篇章整体来说相对重要的特征表示,从而识别出核心篇章单元。在Chinese Discourse Treebank(CDTB)语料库上的实验显示,和最好的基准系统相比,该文的方法在宏平均F1、微平均F1值上均得到了提高。Discourse parsing is the basis of Natural Language Understanding.As one of the important tasks of discourse parsing,nuclearity recognition in Chinese discourse is still an emerging issue.In this paper,we propose a method based on gated memory network(GMN)to recognize nuclearity in Chinese discourse.The method first uses Bi-LSTM and CNN to capture both the remote information and the local information of each discourse units.Then,the two basic discourse units information are merged and a gated unit is created.Finally,gated unit captures relatively important feature representation from the basic chapter unit to identify the Nucleus unit.Experimental results on the Chinese Discourse Treebank(CDTB)show that the proposed method improves the macro-F1 and micro F1 compared to state-of-the-art systems.

关 键 词：篇章分析 主次识别 汉语篇章树库 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]