作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程

• 人工智能及识别技术 • 上一篇    下一篇

一种改进的基于核心句的评价搭配抽取方法

刘 涛1,2,赵 鹏1,2,刘慧婷1,2,纪 霞1,2   

  1.         (1. 安徽大学计算智能与信号处理教育部重点实验室,合肥230039; 2. 安徽大学计算机科学与技术学院,合肥230601)      
  • 收稿日期:2014-01-20 出版日期:2015-02-15 发布日期:2015-02-13
  • 作者简介:刘 涛(1985 - ),男,硕士研究生,主研方向:自然语音处理,文本情感挖掘;赵 鹏、刘慧婷,副教授;纪 霞,讲师、博士。
  • 基金资助:
    国家自然科学基金资助项目(61202227);安徽省自然科学基金资助项目(1408085MF122);安徽大学青年骨干教师培养基金 资助项目(33010017)。

An Improved Method of Evaluation Collocation Extraction Based on Kernel Sentence

LIU Tao 1,2,ZHAO Peng 1,2,LIU Huiting 1,2,JI Xia 1,2   

  1. (1. Key Laboratory of Intelligent Computing & Signal Processing,Ministry of Education,Anhui University,Hefei 230039,China;2. School of Computer Science and Technology,Anhui University,Hefei 230601,China)
  • Received:2014-01-20 Online:2015-02-15 Published:2015-02-13

摘要: 目前主流的评价搭配抽取方法以句法依存分析为基础,由于中文评价文本的不规范性,导致其句法分析结果不稳定,进而影响评价搭配的抽取效果。针对该问题,提出一种改进的基于核心句的评价搭配抽取方法。设计融合核心句和句法依存关系的评价搭配抽取方法,提高评价语句句法分析结果的稳定性,并且在处理复杂的评价语句时,加入对评价对象之间、情感词之间并列关系的分析。实验结果表明,该方法能提高召回率和准确率。

关键词: 核心句, 评价搭配, 语法模板, 并列关系, 依存关系

Abstract: The mainstreaming evaluation collocation extroction methods are based on syntactic dependency parsing. Because the grammar of most Chinese evaluation text is not normative,the syntax analysis result is unstable and affects the result of extracting evaluation collocation. To solve this problem,this paper presents an improved method of extracting evaluation collocation based on kernel sentences,which extracts evaluation collocation by combining kernel sentences and syntactic dependency. This method can significantly improve the stability of the syntax analysis result,and it also can add the analysis of the coordinative relationship among the emotional words and among the opinion targets when dealing with complex sentences. Experimental result exhibits that this method can improve the recall rate and accuracy.

Key words: kernel sentence, evaluation collocation, syntax template, coordinative relationship, dependency relationship

中图分类号: