摘要
复句在书面语中具有举足轻重的地位,如何让计算机正确理解复句是中文信息处理中一个值得重视的问题。现有的分词系统对复句关系词的正确切分与标注上不足以满足对复句进行层次分析和语义分析的需要。建立的分词系统在复句中关系词的切分和标注上做出了必要的改进。
Compound sentences occupy very important status in writing language. How to make computers understand compound sentences correctly is a problem for us to think much of in Chinese information procession. The compound relatives’ segmentation and tagging of the word segmentation systems in existence can’t satisfy the demands of compound sentences hierarchical and semantic analysis. The paper founds a word segmentation system that made some necessary improvement in the segmentation and tagging of compound relatives.
出处
《计算机与数字工程》
2007年第5期43-44,81,共3页
Computer & Digital Engineering
基金
国家重点实验室开放研究基金(编号:SKLSE04-018)资助
湖北省科技公关项目(编号:2005AA101C43)资助
关键词
汉语复句语料库
关系词
分词
Chinese compound sentences corpus,relative words,word segmentation
