基于机械分词与统计学的新词识别研究  

Research on New Word Identification Based on Automatic Segmentation and Statistics

作　　者：梁刚[1] 

机构地区：[1]南京农业大学信息科技学院,江苏210095

出　　处：《情报理论与实践》2005年第5期475-477,共3页Information Studies:Theory & Application

摘　　要：分词是所有中文信息处理工作的基础,也是汉语信息处理的难点之一,如何识别文中出现的新词更成为当前研究的重点之一。本文综合利用几种传统的机械分词方法,加上统计学的方法,提出了一种从文献中抽取新词的新方法。Automatic segmentation is the foundation of Chinese information processing and also one of its difficult points. How to identify the new words in the articles has become one of the research focus at present. This paper puts forward a new method of taking out new words from articles by the comprehensive use of several traditional automatic segmentation methods and the method of statistics.

关 键 词：信息处理技术 自动分词/新词识别 未登录词 统计学 新词 分词 识别 机械 信息处理工作 汉语信息处理 

分 类 号：G254[文化科学—图书馆学]