Study on Discovering the Relationships among Data Resources in DataSpace
|Course||Computer System Architecture|
|Keywords||Dataspace relationships among data resources domain ontology frequent itemsets|
As the most successful software technologies in the field of IT, relational database management system has become the foundation of the commercial, financial and scientific activities in the worldwide。However, with the development of technology, the Computer and Internet are becoming popular, so personal data is expanding rapidly and Web is becoming a huge information-sharing platform, then data management presents some new features:rapid growth, information sharing, diversification of resources, heterogeneous distribution. These new features make the traditional database technology unable to meet the new requirements of data management in new era.Dataspace is a new abstraction for information management which aiming at the challenges of traditional database technology facing. By providing a set of related data management services and mechanisms, the user of dataspace can focus on solving the specific problem, rather than struggling with a large number of related but different underlying data management. Basing on the existing research results about dataspace, this article focuses on organizing data resources, mining relationships among data resources to support the semantic queries. Firstly, a three-tier organizational structure is proposed, which includes physical layer, logical layer and application layer, and the logical layer is divided into logical participant and logical collection, denoted separately as PAD and CKP. Secondly, we mine the direct and indirect relationships among resources to support semantic query. The core idea is as follows: defining the relationships between domain ontology and based on which creating an object-oriented data resources relational graph with direct relationships; mining the indirect relationships among data resources to enhance the data resources relational graph; introducing the confidence of relationships to relational graph. Lastly, with the relational graph we can carry on keyword query, structured query and semantic query effectively.The experiments mainly verify that the relationships among data resources are influenced by logical collection and the query is supported by the relationships among data resources.