摘要
介绍基于文本的本体学习及其层次,分析本体学习中术语获取的主要方法。针对术语获取中存在的问题,在术语形成的经济规律基础上,引入种子概念方法,并利用统计和规则两种方法抽取与种子概念相关的领域术语;证明种子概念方法是一种有效获取领域术语的方法。实验证明少量种子词可以获取大量领域术语,为本体构建提供基础和框架。
This paper introduces texts-based ontology learning and its layers. Then it analyzes the key methods for terms acquisition in ontology learning, In accordance with the problems existing in terms acquisition, the authors discuss the economical law of terms and apply the seed concept method into ontology learning to extract the domain terms relevant to seed concept by statistics and regulations. The experiment shows that seed concept method is an efficient way to acquire domain terms and it may provide basis and frame for ontology construction.
出处
《图书情报工作》
CSSCI
北大核心
2006年第9期18-21,共4页
Library and Information Service
基金
国家自然科学基金"面向自然语言处理的逻辑语义表达与演算模型研究"(项目编号:60173025)成果之一。
关键词
本体学习
文本
术语获取
种子概念
ontology learning text term acquisition seed concept
