【摘要】 专用名词虽然只占中文文章中的词的百分之一到百分之二,但是,如果不对这些专用名词加以处理,将会形成自动分词的错误的大部分。本文首先描述了包括中文姓名辨识的分词方法,然后介绍其实验结果。最后,文章讨论了中文姓名辨识被遗漏和误判的原因,并提出未来的研究方向。更多还原

【Abstract】 The words of non-commonly-used nouns are 1 or 2 percent among the words of Chinese articles, but this will lead to form most of the errors in automatic word segmentation if these words of non-commonly-used nouns are not handled. This paper first describes one automatic word segmenting method including the recoginition of Chinese full name. This paper also introduces its test results. Lastly, this paper discusses the reasons of recoginition of Chinese full name having been lost or wrongly segmented, and puts forward the direction of future research.更多还原