摘要
手工构建本体是一项既费时又费力的工作,为解决此项工作的瓶颈问题,本体自动构建成为当前的一个研究热点和重点。考虑到不同语言描述的本体在本质上是相同的,只是表层的表示符号不同,提出了一种基于本体翻译的领域本体自动构建算法,该算法针对已存在本体中概念标签的不同情况,分别采用不同的统计指标来筛选标签的目标语翻译。通过将一个英文金融本体翻译成中文对算法进行了实验验证,说明了算法的有效性。
Manual ontology construction means large amounts of labor cost and time consuming,to solve the bottleneck problem of this work,the issue of automatic ontology construction becomes the focus of current research,taking into account that ontologies of a particular domain described by different languages are the same in nature despite the varied symbols on the surface level,an algorithm for automatic construction of domain ontology based on ontology translation is proposed.The algorithm selects the target language translation for labels of existing ontology by various statistical measures according to different characteristics of the labels.An experiment on translation of English financial ontology into Chinese is performed to validate the algorithm and validated the effectiveness of the algorithm.
出处
《计算机工程与设计》
CSCD
北大核心
2011年第9期3203-3207,3230,共6页
Computer Engineering and Design
基金
国家自然科学基金项目(90409007)
关键词
语义网
本体
本体翻译
信息熵
译文选择
semantic web
ontology
automatic translation
information entropy
translation selection
