Indexing and Storage Schemes for Keyword-based Query Processing over Semantic Web Data

Metadata and ontology can be used to retrieve related information through the inference mure accurately and simply on the Semantic Web. RDF and RDF Schema are general languages for representing metadata and ontology. An enormous number of keywords on the Semantic Web are very important to make practical applications of the Semantic Web because most users prefer to search with keywords. In this paper, we consider a resource as a unit of query results. And we classily queries with keyword conditions into three patterns and propose indexing techniques for keyword-search considering both metadata and ontology. Our index maintains resources that contain keywords indirectly using conceptual relationships between resources as well as resources that contain keywords directly. So, if user wants to search resources that contain a certain keyword, all resources are retrieved using our keyword index. We propose a structure of table for storing RDF Schema information that is labeled using some simple methods.