Embedding Relational Databases in Object Ontologies: Implementation Issues

In this paper the problem of database embedding in logical knowledge management environment is considered. The approaches to the implementation of database embedding in ontologies are established through their modeling based on object theories. This method focuses on the unified database management within big distributed information systems. In introduction we give the relevance of the investigated problem. At the first part, we point attention on objective theory of databases. According to it, we determine the basic mechanism and the rules of objective model building on a random relational database system. Then we define modeling of external keys of a database through objective theory features. Further we consider the implementation of a query language interpreter «Libretto» on relational databases. Then the general architecture of the «Libretto» system is given. In addition to this, we mention on two ways of implementation mechanisms of the databases which are based on «Libretto». The first one is about implementation Libretto API in the context of DBMS; the latest could be described as a substitution of the translator «Libretto». Later the databases sublanguage «Libretto» is determined on order for implementation method of substitution of the translator «Libretto». This databases sublanguage replies for request generating. Then we analyze translation of elementary requests, transmission ways, transmission of inverse features, transmission of inserted requests, predicates transmission. Finally, we give the comparison of the efficiency of 2 above-mentioned ways, summarize undertaken work and see the further development of these ways