摘要
提出了一种新的汉语自动分词算法,其主要思想是通过前后两次对文章的扫描来解决分词过程中出现的交叉歧义问题,介绍了一种新的有效的字段切分算法,它能够排除类似穷举算法中冗余的单字词的切分可能。
This article presents a new algorithm of automatic Chinese word segmentation. Its main idea is to settle the problem of different meanings under the separating words process by scanning the article two times. And puts forward a new efficient string segmentation algorithm, which can exclude possibilities of redundant single words in other algorithms.
出处
《计算机工程》
CAS
CSCD
北大核心
2004年第16期146-148,共3页
Computer Engineering
关键词
上下文相关
汉语自动分词
分词统计模型
Context relation
Automatic Chinese segmentation
Statistical model of segmentation