Mapping of Description Logic to the Relational Data Model
- PDF / 174,283 Bytes
- 15 Pages / 594 x 792 pts Page_size
- 110 Downloads / 213 Views
NEW TOOLS OF CYBERNETICS, INFORMATICS, COMPUTER ENGINEERING, AND SYSTEMS ANALYSIS MAPPING OF DESCRIPTION LOGIC TO THE RELATIONAL DATA MODEL P. I. Andon,1† V. A. Reznichenko,1‡ and I. S. Chistyakova1††
UDC 004.62
Abstract. General issues of the data integration problem are discussed, including ontology-based data integration in the semantic web. The importance of establishing mappings between ontologies and relational databases is shown, and a brief analytical review of studies in this field is given. The binary relational data model is defined and the mapping of the description logic ALC and its extensions to the binary relational data model is described. Keywords: data integration, ontology, semantic web, description logic, binary relational data model. INTRODUCTION. DEFINITION OF THE DATA INTEGRATION PROBLEM Data integration is one of the most demanded trends in the modern information industry. However, it is not new. The first steps in this field were made in the mid-1970s. At that time, more exact ideas about multilevel architecture of databases (DB), about data models (DM) as a tool of information modeling of the reality, and about mapping between them began to be formed. The main attention was focused on supporting the global scheme, which was called conceptual, for the set of local schemes supporting the same (or different) DM that operate at different nodes under control of one DBMS. Later, the problem was generalized and included creation of data storages, various repositories of information resources and web applications. Currently, integration of heterogeneous data of various information systems is most important in data integration. Over the years of the existence of the Internet space (especially, the last decade), it has accumulated a great amount of information, and this amount increases in geometrical progression. This generates many important problems related to the use and storage of data in information field. Huge amount of data in heterogeneous sources represent information by different ways and have different functional purposes. Integration and sharing of information from the set of such data sources is a challenge, permanently important throughout the last decades. Let us consider several reasons why the heterogeneity occurs. Various Data Models. As is generally known, 75 to 90% of information is stored in relational DB (RDB). However, the other 10–25% include the considerable amount of data stored in structures that are defined by absolutely other DM with their specific semantics. Different Methods of Data Storage (Files, DB, Warehouses, etc.). Physical organization of information storage creates additional obstacles for its use. Data integration should provide a uniform logical format of data organization so that irrespective of the way of their physical storage, end user would have a uniform mechanism of access to the contents.
1
Institute of Software Systems, National Academy of Sciences of Ukraine, Kyiv, Ukraine, †[email protected]; [email protected]; ††[email protected].
Data Loading...
