摘要
针对大量领域信息资源缺乏有效内容分析这一问题,设计、开发了通过构建领域本体来充分发掘信息资源之间内在关系,实现信息资源增值服务的原型系统.首先构造了服装知识本体CO(Costume Ontology)用来表示服装领域知识.利用CO对被管信息资源进行语义描述,在描述数据库的基础上,实现基于语义的信息检索.讨论了服装知识本体的建模过程,本体的形式化表示以及存储方式.实验系统采用J2EE架构,开发了存储组件、语义查询接口组件、语义分析组件和语义推理组件.通过使用资源描述框架RDF,实现被管信息资源到知识本体层的映射.通过语义分析和语义推理,可以充分利用信息资源之间的关系实现相关信息资源检索与语义融合.
To solve the lacking content based analysis problem on volumes of domain information resource, a domain ontology based information resource management platform trial system was designed and implemented. Costume ontology(CO) was created to represent domain concepts and the relationship between concepts. CO was used to construct the description database of managed information resource to calculate the relationship between these resource and to add value to them. CO based semantic query on the description database was implemented. The costume domain ontology modeling, ontology formal representation and storage schema were discussed. The trial system took J2EE architecture which including storage component, semantic query interface component, semantic analysis component and semantic inference component. Using resource description framework( RDF), the meaning of the managed resource was mapped to the ontology layer(CO). By semantic analysis and semantic inference, the relationship among resource could be fully used to implement the relevant information resource retrieval and semantic fusion.
出处
《北京航空航天大学学报》
EI
CAS
CSCD
北大核心
2005年第11期1245-1249,共5页
Journal of Beijing University of Aeronautics and Astronautics
基金
国家863计划基金资助项目(2003AA414016)