摘要
基本名词短语的识别在自然语言信息处理领域具有重要作用。本文首先从语言学的角度提出了汉语基本名词短语的概念，然后从语言信息处理的角度将用于基本名词短语识别的知识分为两部分，即表示基本名词短语句法组成的基本结构模板（静态知识）与表示基本名词短语出现的上下文环境特征的转换规则（动态知识）。在此基础上设计了一种基于转换的基本名词短语识别模型，该模型可同时结合这两类知识识别基本名词短语。
It is important to recognize the baseNP in the field of natural language processing. At first, the paper defines Chinese baseNP from the linguistic standpoint. Then the knowledge which is essential for baseNP recognition is analyzed from the standpoint of automatic language information processing. The recognition knowledge includes the basic construction templates which specify the syntactic composition of baseNPs(static knowledge) and the context sensitive transformative rules(dynamic knowledge) which reflect the context features. Based on the above knowledge, a transformation based model for recognizing Chinese baseNP is put forward, which incorporates the static knowledge and the dynamic knowledge into an organic whole to recognize the baseNPs in Chinese texts. The experiment shows a satisfatory precision and recall.
出处
《中文信息学报》
CSCD
北大核心
1999年第2期1-7,39,共8页
Journal of Chinese Information Processing
关键词
自然语言处理
知识获取
名词短语识别
汉语
Natural Language Processing Knowledge Acquisition Corpus Noun Phrase
