【摘要】 [目的 /意义]针对当前我国高校机构知识库研究与应用离真正的开放互连、知识集成、知识检索还有较大差距的问题,提出一种面向关联数据的机构知识库构建方法,该方法覆盖机构知识库构建的整个过程,并支持机构知识的资源整合、语义检索、知识推理和关联数据发布,为各高校机构知识库建设提供参考。[方法/过程]调研和剖析国内外典型的机构知识库,将机构知识库构建方法分为5个步骤,包括:1评价和选择有价值的信息资源;2建立机构知识库概念模型;3进行语义描述和关联设置;4发现并创建关联化数据环境下的语义推理规则;5设计系统架构,并通过西南科技大学机构知识库构建实例来验证该方法的有效性。[结果 /结论]实例验证表明,面向关联数据的机构知识库构建方法具有合理性和有效性,起到了降低机构知识库开发难度和缩短开发周期的作用,对于其他机构知识库的建设具有参考价值。更多还原

【Abstract】 [Purpose/significance]This paper investigates the problem of college institutional repository’s( IR) deficiency of capacity of open linking,knowledge integration and knowledge retrieval,puts forward a linked data-oriented construction method of academic IR which covers all the process and supports knowledge integration,semantic retrieval,knowledge reasoning and linked data release,providing reference to subsequent institutional repository construction. [Method /process]Investigates and analyzes typical institutional repositories at home and abroad,divides the method of IR construction into five steps: 1 Evaluating and selecting valuable information resources; 2 Building conceptual model of IR; 3 Making semantic annotation for instance data and interconnecting it with external open dataset at the same time; 4 Discovering and setting semantic inference rules under linked open data environment; 5 Designing system structure. The method’s availability is tested and verified from the practice of SWUST( Southwest University of Science and Technology) IR construction. [Result /conclusion]Practice verifies that linked data-oriented construction method of IR is rational and practical and takes the effect of reducing difficulty and shortening period of IR construction. The method is worthy of extending to other IR’s construction.更多还原