首页 | 本学科首页   官方微博 | 高级检索  
     

平行推理机制:一种隐式篇章关系检测方法
引用本文:洪宇,严为绒,车婷婷,梁颖红,姚建民,朱巧明,周国栋.平行推理机制:一种隐式篇章关系检测方法[J].软件学报,2014,25(11):2528-2555.
作者姓名:洪宇  严为绒  车婷婷  梁颖红  姚建民  朱巧明  周国栋
作者单位:苏州大学 计算机科学与技术学院, 江苏 苏州 215006;苏州大学 计算机科学与技术学院, 江苏 苏州 215006;苏州大学 计算机科学与技术学院, 江苏 苏州 215006;苏州职业大学 计算机系, 江苏 苏州 215104;苏州大学 计算机科学与技术学院, 江苏 苏州 215006;苏州大学 计算机科学与技术学院, 江苏 苏州 215006;苏州大学 计算机科学与技术学院, 江苏 苏州 215006
基金项目:国家自然科学基金(61373097, 61272259, 61272260, 90920004);教育部高等学校博士学科点专项科研基金(200932 01110006);苏州市国际科技合作计划(SH201212)
摘    要:篇章是论元经过语义关联和结构化组织形成的自然语言文体.篇章分析研究的核心任务之一是解释论元的语义关系,其中,显式关系因具有直观线索而易于检测,目前检测精度高达90%;相对而言,隐式关系因缺乏直观线索而难于检测,目前精度仅约40%.针对这一问题,基于一种"论元平行则关系平行"的假设,并利用显式篇章关系易于检测的特点,通过平行论元的识别与平行关系的消歧,实现了一种显式关系平行推理隐式关系的隐式篇章关系检测方法.利用标准宾州篇章关系树库(Penn discourse Tree Bank,简称PDTB)对这一检测方法进行评测,结果显示,精确率提升达17.26%.

关 键 词:篇章  隐式篇章关系  关系检测  平行线索  篇章关系消歧
收稿时间:2013/5/14 0:00:00
修稿时间:2013/12/4 0:00:00

Parallel Inference: A Method of Implicit Discourse Relation Detection
HONG Yu,YAN Wei-Rong,CHE Ting-Ting,LIANG Ying-Hong,YAO Jian-Min,ZHU Qiao-Ming and ZHOU Guo-Dong.Parallel Inference: A Method of Implicit Discourse Relation Detection[J].Journal of Software,2014,25(11):2528-2555.
Authors:HONG Yu  YAN Wei-Rong  CHE Ting-Ting  LIANG Ying-Hong  YAO Jian-Min  ZHU Qiao-Ming and ZHOU Guo-Dong
Affiliation:School of Computer Science and Technology, Soochow University, Suzhou 215006, China;School of Computer Science and Technology, Soochow University, Suzhou 215006, China;School of Computer Science and Technology, Soochow University, Suzhou 215006, China;Department of Computer, Soochow Vocational University, Suzhou 215104, China;School of Computer Science and Technology, Soochow University, Suzhou 215006, China;School of Computer Science and Technology, Soochow University, Suzhou 215006, China;School of Computer Science and Technology, Soochow University, Suzhou 215006, China
Abstract:Discourse is a literary form that consists of semantically-related and well-structured arguments. One of the key tasks of discourse analysis is to resolve semantic relationship between arguments. Explicit relation is easy to detect with an accuracy of nearly 90% because of its direct cues. In contrast, implicit relation is difficult to detect with only an accuracy of nearly 40% since it has no direct cues. To solve the problem, paper proposes a hypothesis that parallel arguments normally have the consistent semantic relations. Based on the hypothesis and by utilizing the characteristics that explicit relation is easy to detect, the paper implements a method of implicit relation detection which uses explicit relation to infer implicit relation among parallel arguments. We evaluate the method on the standard penn discourse Treebank (PDTB). The experimental results show an improvement of 17.26%.
Keywords:discourse  implicit discourse relation  relation detection  parallel cue  discourse relation disambiguation
本文献已被 CNKI 等数据库收录!
点击此处可从《软件学报》浏览原始摘要信息
点击此处可从《软件学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号