Enhancement of Query Execution Time in SPARQL Query Processing

Abstract

During couples of decades, the popularity of Semantic Web is increased in both academic and industry. RDF is a data model standardized by W3C for the Semantic Web and a query language used for retrieving the RDF data is SPARQL query. Most of the real-world applications has widely used RDF as a standardized data format, so that efficient data storage and query processing techniques for the RDF data need to be considered. In this paper, an indexing structure and query processing technique are developed to enhance the execution time of three different types of SPARQL queries such as conjunctive, optional, and alternative queries. As the RDF data is formed in graph structure, the indices are constructed by extracting the subgraphs around each vertex based on the edges nature of the vertex. Although the size of query is large especially in conjunctive query, our proposed graph-based index and query processing method can help to save the query execution time. Experimental results show that our proposed method works well for all three different types of SPARQL queries and outpaces the competition at processing the conjunctive queries.