计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (24): 153-155.DOI: 10.3778/j.issn.1002-8331.2009.24.045

• 数据库、信息处理 • 上一篇    下一篇

基于汉语情感词表的句子情感倾向分类研究

王素格1,杨安娜1,李德玉2   

  1. 1.山西大学 数学科学学院,太原 030006
    2.山西大学 计算机与信息技术学院,太原 030006
  • 收稿日期:2008-05-09 修回日期:2008-07-28 出版日期:2009-08-21 发布日期:2009-08-21
  • 通讯作者: 王素格

Research on sentence sentiment classification based on Chinese sentiment word table

WANG Su-ge1,YANG An-na1,LI De-yu2   

  1. 1.School of Mathematics Science,Shanxi University,Taiyuan 030006,China
    2.School of Computer & Information Technology,Shanxi University,Taiyuan 030006,China
  • Received:2008-05-09 Revised:2008-07-28 Online:2009-08-21 Published:2009-08-21
  • Contact: WANG Su-ge

摘要: 提出了一种基于汉语情感词词表的加权线性组合的句子情感分类方法。该方法通过已有的五种资源构建了中文情感词词表,并采用加权线性组合的句子情感分类方法对句子进行情感类别判断。实验结果表明,直接利用词汇语言粒度的句子情感分类综合F值为78.62%,若加入了否定短语语言粒度后,句子情感分类的综合F值提高了4.14%。

关键词: 情感词表, 权线性组合, 子情感分类

Abstract: This paper presents the weighted linear combination method for the sentence sentiment classification based on Chinese sentiment word table.In proposed method,firstly,Chinese sentiment word table is constructed by using five existing dictionaries,secondly,automatically identifying method is explored for the sentence sentiment classification using the weighted linear combination method.The experiment results indicate that the F value of sentence sentiment classification with word language granularity reaches 78.2%,and adding the negative phrase to language granularity,the F value of sentence sentiment classification has increased by 4.14%.

Key words: sentiment word table, eighted linear combination, entence sentiment classification

中图分类号: