An Approach to Constructing an Integrated Ontology for Integrating Object-Oriented Data Models

Abstract

Semantically determining similarities and differences between data from multiple data sources plays an important role in the process of data integration. The similarities and differences identify the degree of relationships and are used for constructing the integrated data model. In structured data such as object-oriented database, their data models reveal only structure-based semantics without other additional
semantics. This paper proposes an approach to integrate two object-oriented data models based on the integration of ontologies that describe their semantics. The ontology for an object model describes semantics based on the model structure
and additional semantics, i.e. class and attribute semantics, and synonym and hypernym lists. The ontologies are analysed to devise an integrated data model that will be used to facilitate query on the integrated data.