The Sensitive Feature Selection for Chinese Phrases

摘要：

Traditional Chinese text chunking approach is to identify many phrases using only one model, and the same features are used to identify these phrases too. So the helpful features of each phrase are ignored. In fact, different phrases have different helpful features. In this paper, the concept of sensitive feature is proposed, and the sensitive features of seven Chinese types of phrases are selected by dynamic comparison strategy. Through testing on the Multi-agent Chinese chunking model, the selected sensitive features are effective.

关键词： feature phrase shallow parsing

作者: Ying-Hong Liang Jin-xiang Li De-fu Zhou

作者单位: JiangSu Provisee Support Software Engineering R&D Center for Modern Information Technology Applicati Suzhou Vocational University School of Computer Engineering Suzhou, China

会议类型: 国际会议

会议名称: 2010 3rd IEEE International Conference on Computer Science and Information Technology(第三届IEEE计算机科学与信息技术国际会议 ICCSIT 2010)

会议地点: 成都

会议语种:英文

页码: 138-141

在线出版日期: 2010-07-07（万方平台首次上网日期，不代表论文的发表时间）