A Path-based Relational RDF Database

Matono, A., Amagasa, T., Yoshikawa, M. and Uemura, S.

    We propose a path-based scheme for storage and retrieval of RDF data using a relational database. The Semantic Web is much anticipated as the nextgeneration web where high-level processing of web resources are enabled by underlying metadata described in RDF format. A typical application of RDF is to describe ontologies or dictionaries, but in such applications, the size of RDF data is large. As largesize RDF data are emerging and their number is increasing, RDF databases that can manage large-size RDF data are becoming ever more important. To date, some RDF databases have already been proposed; however, they have critical problems: the performance of path queries is insufficient and they cannot discriminate between schema data and instance data. In this paper, as a solution to these problems, we propose a path-based relation RDF database. In our approach, we first divide the RDF graph into subgraphs, and then store each subgraph by applicable techniques into distinct relational tables. More precisely, all classes and properties are extracted from RDF schema data, and all resources are also extracted from RDF data. Each is assigned an identifier and a path expression, and stored in corresponding relational table. Because our proposed scheme retains schema information and path expressions of each resource, unlike most conventional RDF databases, it is possible to process path-based queries efficiently and store RDF instance data without schema information. The effectiveness of this approach is demonstrated through several experiments.
Cite as: Matono, A., Amagasa, T., Yoshikawa, M. and Uemura, S. (2005). A Path-based Relational RDF Database. In Proc. Sixteenth Australasian Database Conference (ADC2005), Newcastle, Australia. CRPIT, 39. Williams, H. E. and Dobbie, G., Eds. ACS. 95-103.
pdf (from crpit.com) pdf (local if available) BibTeX EndNote GS