摘要
为解决文本数据挖掘等知识体系欠发育领域之本体的构建问题,先构建领域本体的最基本概念词集,利用样本库检测与优化基本概念集,利用样本库构造基本概念词的上下文名词与动词,从而筛选出基本概念词的相关名词,构造其语义环境,通过比对语义环境确定同义词、近义词、反义词。将相关名词作为下一代概念词,迭代计算直至构造整个领域本体,并设计了具体算法,证实了该方法的可行性。
To obtain the application domain ontology whose concept system is not growthy,a kernel concept word set is proposed and are refined according to the sample documents,to find the nouns and verbs in context of the kernel concept word,and to select the core relation word and to form the semantic surroundings.Finally,to obtain the synonymy word,approximate word and antonym by matching the semantic surroundings,the selected relation word are used for the kernel concept word of next iterativehess,and the algorithm is developed.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第24期153-155,共3页
Computer Engineering and Applications
关键词
本体
本体构建
概念体系
ontology
ontobuilder
concept system
