基于描述逻辑的语义Web本体研究

【作者】 魏榴花；

【导师】 沈洁；

【作者基本信息】 扬州大学 ， 计算机应用技术， 2008， 硕士

【摘要】 1998年,Tim Berners-Lee提出了“语义Web”的概念,语义Web研究的主要目的就是扩展当前的WWW,使得网络中信息都是具有语义的,便于人和计算机之间的交互合作。语义Web的兴起促进来本体技术的发展,本体作为语义Web的基石,而描述逻辑作为语义Web的一个逻辑基础,对它们的研究具有重要的理论价值和现实意义。本文的主要工作如下。1.面向语义Web领域本体的建立与形式化研究。The World Wide Web Consortium (W3C)正在制定的OWL DL是一种面向语义Web的知识表示标记语言,本文主要以大学本体为例,介绍领域本体建立和形式化研究的一般方法,首先采用NKI(National Knowledge Infrastructure)语言描述了大学领域本体的类,同时规范了它的属性和关系,然后详细分析了OWL DL和描述逻辑之间的对应关系,并通过具体的大学领域本体的实例用描述逻辑的语义解释了OWL DL的各个元素,这样在描述逻辑领域的研究成果就能应用到OWL DL上来,为OWL DL的知识表示和推理问题奠定来一定的基础。2. Web商务智能语义平台。本文以Web商务智能领域为研究对象,建立了该领域的语义平台,具体讲述了它的Web数据资源模块、知识库模块和用户模块,具体分析了本体的演化过程,因为Web商务信息具有明显的时效性和上下文相关性,所以本文也同时建立了Web商务智能领域的时序知识模型,并确定了时间本体的原子概念、关系和公理,实现了基于描述逻辑的概念定义和公理的确定,奠定了形式推理的基础。3.语义Web本体不一致性研究。本体在“语义Web”中起到了至关重要的作用,它通过定义精确的共享术语,以提供某一特定领域可重用的知识。但这些知识并不是静态的,而是随着时间的推移不断演化。本体在演化发展的过程中,不可避免地会产生本体知识库的不一致现象,本文解决了在本体演化后本体知识库产生的不一致性的处理问题,例如对不一致的检验,不一致性的测量,后者对本体知识库的修复具有良好的参考价值。在不一致的测量计算过程中,为了简化计算,我们采用基于结构划分和最小不一致集的方法来对不一致值的计算进行优化,这样就能大大地减少了不一致值计算时间。更多还原

【Abstract】 In 1998, Tim Berners-Lee puts forward concept of“Semantic Web”, the major research aim is to expand the current WWW and make information in the network semantic ,machine-understandable and machine-processable, which is helpful for the interaction and cooperation between humans and computers. The springing up of semantic Web fosters the development of Ontology technology. As Ontology is the foundation of semantic Web, while description logic is logic basis of semantic Web, they have become important content for its research and development. The major work is as below.1. The construction of domain Ontology and its formal inference research. OWL DL, made by W3C, is a kind of knowledge representation and annotation language oriented towards semantic Web. In this paper, we take university Ontology for example to introduce the general method for construction and formalization of domain Ontology, firstly we describe classes of university domain by NKI(National Knowledge Infrastructure) language and specify its properties, relations, then the corresponding relation between OWL DL and description logic is analyzed in detail here, and by the concrete example of university Ontology, every element of OWL DL is interpreted by description logic semantic, by this way, the research achievements in description logic can be applied on OWL DL, which lays basis for knowledge representation and inference problems.2. Web semantic platform of business intelligence. Here we take Web business intelligence domain as research objects, we establish semantic platform of such domain, analyze its Web data source functional module, knowledge base one and user one and analyze evolution process of Ontology in detail. For the reason that Web business information has apparent time efficiency and context-sensitivity, here we establish time sequence knowledge model of Web business intelligence domain too, determine atomic concepts, relations and axioms of time Ontology, realize the aim that determine concept definition and axioms based on description logic, therefore, lay basis for formal inference.3. Inconsistency study of Semantic Web Ontology. Ontology plays a very important role in semantic Web, by defining precise shared terms, it offers reusable knowledge of a specific domain. But the knowledge is not static, it evolves continously with the passage of time. In the process of Ontology evolution and development, inevitably Ontology will be inconsistent. Here we solve the inconsistent problems that occur after evolution on Ontology knowledge base, such as inconsistency test and inconsistency measurement, the latter is of nice reference value for the repair of Ontology knowledge base. In the process of inconsistency computation, for the aim of computation simplification, we apply structure-division based method and minimal inconsistency subset one to optimize the computation of inconsistency, by the way, inconsistency computation can be reduced greatly.更多还原