厦门大学学报（自然科学版）2017，Vol.56Issue(4)：576-583,8.DOI:10.6043/j.issn.0438-0479.201701010
基于混合树结构神经网络的隐式篇章关系识别
A Hybrid Tree Structured Neural Network for Implicit Discourse Relation Recognition
摘要
Abstract
The most critural challenge of implicit discourse relation recognition lies in how to represent the semantic information of each discourse argument.However,the semantic value of the sentence is mainly decided by its specific information focus in linguistics.Therefore,the discourse relation may mostly depend on links between information focuses.Intuitively,we cannot give equal treatment to every phrase branches during composition up the syntactic parse tree.To resolve the problem,we introduce the tree-structured long short-term memory(Tree-LSTM) network to selectively incorporate information from each child to compute the distributed semantic representation of two arguments.Consequently,it can emphasize those informative predicative branches that indicate the "focus" of a sentence.Then the neural tensor network(NTN)is used to predict the semantic correlation between these two discourse arguments across multiple dimensions.Experimental results on PDTB corpus show that our model has achieved some improvement on the task of discourse relation recognition.关键词
隐式篇章关系识别/信息焦点/树状长短时记忆网络/神经张量网Key words
implicit discourse relation recognition/specific information/tree-structured long short-term memory(Tree-LSTM)/neural tensor network(NTN)分类
信息技术与安全科学引用本文复制引用
郑江龙,陈锦秀..基于混合树结构神经网络的隐式篇章关系识别[J].厦门大学学报（自然科学版）,2017,56(4):576-583,8.基金项目
国家自然科学基金(60803078) （60803078）
福建省自然科学基金(2010J01351) （2010J01351）
教育部海外留学回国人员科研启动基金 （）
