On ontology based data integration: Problems and solutions

Abstract

© Published under licence by IOP Publishing Ltd. In the article essential problems of integrating heterogeneous data, arising in development of corporate databases intellectual access systems, are considered. In addition to the common structural problems, caused by variety of data organization, special attention is paid to the less obvious linguistic problems, caused by differences in data notation. A unified approach to overcoming such problems by sequential application of explicit definition of semantics, is described. This approach was tested in development of an intelligent search system for the TATNEFT oil-producing corporation; the system implementation showed high relevance of search results together with an adequate reactivity.