Studies On Construction And Storage Of RDF(S) Based On Databases

The Semantic Web is an extension of the current Web, in which Web resources are given computer-understandabale semantics, better enabling computers and people to work in cooperation. The core of the Semantic Web is to add semantic metadata for the Web resources that can be understood by computers, so that the entire Web becomes a universal medium of exchanging information. The Resource Description Framework RDF and its schema RDF Schema (collectively referred to as RDF(S)) are a recommendation of resource annotation in the Semantic Web. RDF(S) plays important roles in the Semantic Web, and has attracted considerable attention.The success and proliferation of the Semantic Web largely depends on the management of RDF(S) data, including construction, storage, query, and etc. From the view of construction, many application domains contain much valuable information, and in order to achieve reusability and a high level of interoperability of knowledge, how to extract RDF(S) from the existing sources has become a research topic of the Semantic Web. From the view of storage, with the development of the Semantic Web, much RDF(S) data came into being successively, and thus how to store them effectively has become more and more important. Databases are the main way of storing and managing data resources in many application domains. Therefore, the storage of RDF(S) data supported by the database techniques has become a very important research topic of the Semantic Web. From the view of query, with the emergence of large amounts of RDF(S) data, how to query efficiently the large-scale RDF(S) data has attracted much attention. Lots of the existing RDF(S) tools (e.g., Jena and Sesame) manage the RDF(S) data with the relational database systems, and thus some mature database techniques of querying and optimizing could be used to query the RDF(S) data sets. However, it should be noted that the standard query language of RDF(S) is SPARQL (Simple Protocol and RDF Query Language), and relational database is SQL. In this case, the investigation of efficient query methods for RDF(S) based on the transformation of SPARQL into SQL has become a very important research isssue in the area of RDF(S) data management.In recent years, many researches have devoted to investigate the related techniques of RDF(S) data management, such as construction, storage, and query. It should be pointed out that, the existing RDF(S) data management technologies are often difficult to meet the needs of the real-world applications, and the improvement and research about these techniques actually have been continued, so as to promote their progress and development. To this end, the construction, storage, and query techniques of RDF(S) data management are investigated in depth, and a RDF(S) framework from construction, storage, to query is developed, and some concrete solutions and methods are presented. Our main contributions are briefly summarized as follows:(1) To deal with the problem of construction of RDF(S), a complete construction approach from UML to RDF(S) is proposed. The UML (Unified Modeling Language) has been the de facto standard and extensively used in the design phase of database modeling. Therefore, how to construct RDF(S) from UML becomes an important issue to be solved in the knowledge management of the Semantic Web. To this end, formal definitions of UML and RDF(S) are presented. Then, by comparing and analyzing the characteristics of UML and RDF(S), an approach for constructing RDF(S) from UML is proposed, some detailed construction rules are given, a full construction example is provided, and the analyses and discussions about the approach are done. Finally, based on the proposed approach, a prototype construction tool is implemented, and the experiment shows that the approach and the tool are feasible.(2) To deal with the problem of storage of RDF(S), a multiple storage model of RDF(S) based on relational database is proposed. Firstly, by analyzing the characteristics of the RDF(S) instance data and schema semantic information in depth, an overall storage framework of RDF(S) in relational database is provided. Then, some detailed storage rules and algorithm, and a storage example are given. Also, the correctness of the storage approach is discussed and proved. Finally, based on the proposed storage approach, a prototype storage tool is implemented, and the experiment shows that the approach and the tool are feasible. Also, some comparisons and analyses with the existing methods are done.(3) To deal with the problem of query of RDF(S), an approach of translation of SPARQL into SQL is proposed to achieve efficient RDF querying. Firstly, after the in-depth studies on the query mechanism of SPARQL and the translation of SPARQL triple graph patterns to SQL queries, a SPARQL structure tree based on the graph patterns is created, and some translation algorithms related to the traverse nodes when pre-order traversing the structure tree are given. Then, the translation rules of SPARQL queries to SQL queries are provided. Finally, an experimental system called SPARQLtoSQL is implemented. The experimental results are analyzed which demonstrate the feasibility, the correctness and the effectiveness of the system, and some comparisons with the existing methods are made.