首页 | 官方网站   微博 | 高级检索  
     

基于搭配模式的汉语词性标注规则的获取方法
引用本文:王素格,张永奎.基于搭配模式的汉语词性标注规则的获取方法[J].计算机工程与应用,2001,37(5):56-58.
作者姓名:王素格  张永奎
作者单位:山西大学计算机科学系
基金项目:国家自然科学基金项目!(69575011),国家“863”项目!(863-306-ZT03-03-1),山西省自然科学基金项目!(991035),山西省归国人
摘    要:文章介绍了一种基于搭配模式的汉语词性标注规则的获取方法。该方法从已标注了词性的语料库中自动获取候选搭配模式规则,然后根据可信度从候选规则中选择出大于某阈值的规则,再通过不断测试新语料来完善规则。将获取的规则用于汉语的词性标注,使标注的正确率得到了明显提高。

关 键 词:词性标注  语料库  搭配模式  词性标注规则
修稿时间:2000年10月1日

A Acquisitive Method of Chinese Pos Tagging Rules Based-on Collocation Mode
WANG Suge,ZHANG Yongkui.A Acquisitive Method of Chinese Pos Tagging Rules Based-on Collocation Mode[J].Computer Engineering and Applications,2001,37(5):56-58.
Authors:WANG Suge  ZHANG Yongkui
Abstract:In this paper,a kind of acquisitive method for Chinese POS tagging rules based on collocation mode,is introduced. It can obtain candidate rules of collocation mode from POS-tagged corpus,choose those rules,whose threshold is above some value,from candidate rules according to confidence degree,and then expand the rules base by testing new corpus constantly. In the experiment,Authors applied the rules to tagging Part Of Speech for Chinese,the precision has greatly improved.
Keywords:: part-of-speech,corpus,collocation mode,POS tagging rule
本文献已被 CNKI 维普 万方数据 等数据库收录!
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-23

京公网安备 11010802026262号