一种面向结构化数据源的语义标注和挖掘方法  

A method of semantic annotation and mining for structured data sources

作　　者：李璞[1,2] 肖宝[3] 王惠 

机构地区：[1]郑州轻工业学院软件学院,河南郑州450000 [2]华南师范大学计算机学院,广东广州510631 [3]钦州学院电子与信息工程学院,广西钦州535000 [4]国家新闻出版广电总局542台,北京124450

出　　处：《南阳师范学院学报》2016年第6期22-26,45,共6页Journal of Nanyang Normal University

基　　金：河南省基础与前沿技术研究计划项目(102300410110);广西高校中青年教师基础能力提升项目(KY2016LX431)

摘　　要：面对海量数据,结构化数据源作为一种新的知识表示和发布形式,可以提供更加精确和易于理解的语义.为了进一步发现和获取结构化数据源中的潜在知识,提出了一种新的语义标注和推理方法,用来挖掘隐含语义.具体而言,鉴于RDF已成为描述结构化数据源的主要工具,本文首先分析了RDF三元组中二元关系的语义属性,在此基础上设定了一系列语义标注和推理规则,并给出了一种通用的面向结构化数据源的语义增强型扩展方案.最后通过测试,验证了这种语义标注和挖掘方法的有效性.Faced with massive data,structured data sources,as a new form of knowledge representation and publishing,can provide a more precise and comprehensible semantics. To further discover the implicit knowledge contained in structured data source,in this paper,a new semantic annotation and reasoning approach was presented which can discover the latent semantics. Particularly,in view of RDF has become a major tool for describing structured data sources,the properties of the binary relation in RDF triple were analyzed. Based on this,some rules for semantic annotation and reasoning was designed. Further,a novel general semantically-enhanced extension scheme for structured data sources was proposed. Lastly,the test results proved the feasibility of this semantic annotation and mining method.

关 键 词：语义标注 结构化数据源 语义增强型扩展 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]