Study On Discovering The Relationships And Semantic Query Among Data Resources In DataSpace

After years of development, database technology has been one of the most successful technologies in IT field. The development of global data management in business, government, finance and so on has been promoted. The traditional relational database has been unable to cope with three typical characteristics of contemporary data. The first one is mass data. The second one is information sharing. The third one is the diversified types of data. And the last one is the rapid data updating. According to these characteristics, the traditional database has been unable to meet the growing needs of current information development.Dataspaces are an abstraction in data management that aim to overcome some of problem encountered in data integration system. The aim is to reduce the effort required to set up a data integration system by relying on existing matching and mapping generation techniques, and to improve the system in pay-as-you-go fashion as it is used. Traditionally, data integration and data exchange systems have aimed to offer many of the purported services of dataspace systems.Firstly, in this thesis, dataspace is divided into the object model layer, the base layer, the resource layer, the analysis layer, and the management layer based on the analysis of the historical dataspace management system architecture. Secondly, the SPORD model is brought forward to represent the data organization on the resource layer. Then a multi-dimensional association rule discovery algorithm based on MDFP-tree is brought forward for the relationship between the data nodes on the resource layer in dataspace. Thirdly, the content query, structured query and semantic query based on Trie tree is brought forward for the query processing in dataspace. The semantic query processing through MS-SteinierTree algorithm and top-k order of the results is emphatically introduced. Finally, using C#and VS2005as the experimental platform, the MDFP-tree algorithm and the MS-SteinierTree algorithm is tested on performance and feasibility. The experimental results show that the MDFP-tree algorithm and the MS-SteinierTree algorithm are better than the similar algorithms in the performance, and the models and algorithms in this thesis are demonstrated to be feasible.