摘要
利用分词和词性标注的信息,针对未登录词中的中文机构名的组成进行了深入的分析,总结出机构名的内部组成特点,提出了基于模板匹配的中文机构名识别的方法。给出了中文机构名的模板和识别过程,介绍了机构名出现的边界条件。在开放测试中,中文机构名识别的精确率和召回率分别为92.1%和72.81%,取得了较好的识别结果。
After deeply analyzing of the components of Chinese organization name, this paper summarizes its characteristics with help of segmentation and tagging information. Then it propose a method for Chinese organization name recognition based on template matching. This paper presents the template and recognizing procedure, and introduces the boundary condition of Chinese organization name occurring. The precision and recall of Chinese organization name recognition are 92.1% and 72.81% in the opened-test.
出处
《信息技术》
2008年第6期97-99,共3页
Information Technology
关键词
未登录词
中文机构名识别
模版匹配
out-of-vocabulary
Chinese organization name recognition
template matching
