基于维基百科的语义知识库及其构建方法研究

摘要

维基百科(Wikipedia)是规模最大的在线网络百科全书之一,采用群体在线合作编辑的Wiki机制,具有质量高、覆盖广、实时演化和半结构化等特点,是用来构建语义知识库的优质语料来源.分析了维基百科语料库的基本情况,综述了目前基于维基百科所构建的多种语义知识库及其概念抽取和关系抽取方法,讨论了各类方法的优缺点、开放问题和可能的研究方向.%Wikipedia is one of the largest online encyclopedias, which takes the mechanism of online cooperating editing. It is high-quantity, wide-coverage, evolving and semi-structural, and has become a well corpus of semantic knowledge bases. This paper analysed the statics and mechanism of Wikipedia, summaried the semantic knowledge bases and their construction techniques, concluding concepts and relationships mining, at last discussed the open problems in knowledge construction based on Wikipedia.