Resource Description Framework (RDF), a standard language model for representing semantic data, becomes more important to retrieve and exchange information as the Semantic Web gets more viable. Because efficient management of RDF data is one of the key research issues in Semantic Web, there are many RDF management system proposals with data storage architectures and query processing algorithms to retrieve data. However, most of approaches require many join operations based on the complexity of the queries and produce a lot of unnecessary intermediate results to processing SPARQL queries. The problem gets more serious with a huge amount of RDF data source.
In this thesis, we propose a new structural index in an efficient manner with a query optimizer for processing query for a large scale RDF data without any join operations using structural indexing approach. In this approach, we process the query with it execution plan to reduce the useless intermediate data, hence improve performance of querying RDF data.
The empirical experiment results show that our proposed system outperforms a system such as Jena that uses conventional approach by reducing unnecessary intermediate results in query processing.
