摘要
提出一种利用翻译词对信息快速构建知识组织系统中术语之间同义关系的方法:利用单词条的翻译数量、翻译共现强度等作为基本输入,通过简单布尔判断、条件互信息或向量空间模型余弦相似度来计算术语之间的相似度,进而发现并构建同义关系;采用准确率和丰富率来评价构建效果,通过在新能源汽车领域进行实验,发现该方法不但速度快,而且在翻译信息比较丰富的条件下,对于中文同义关系构建有较好效果。
With tile translation words pairs information, this paper proposes a rapid construction method of synonym ,elationship between terms in the knowledge organization systems. It makes the amount of translation entries and the co - occurrence intensity of tra,lslation as the basic inputs. The similarity between terms is calculated by simple Boolean judge, conditional mutual intbrmation and cosine similarity of vector space model. The synonym relationships are discovered and constructed. Precision and richness are used as the indicators to evaluate the method effects. With an experiment in new energy vehicle domain, this paper finds that the method＇ s speed is very fast, and the construction effects are better in the condition of enough translations information.
出处
《图书情报工作》
CSSCI
北大核心
2013年第8期109-113,共5页
Library and Information Service
基金
国家自然科学基金项目“面向特定情报分析应用的知识组织系统快速构建关键问题研究”(项目编号:71203208)
国家“十二五”科技支撑计划课题“面向外文科技文献信息的超级科技词表和本体建设”(项目编号:2011BAH10B01)研究成果之一
关键词
知识组织系统
汉语科技词系统
同义关系
快速构建方法
翻译信息
relationshipknowledge organization systemconstruction method translationChinese scientific and technical vocabulary system synonym
