期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：55
期号：1
出版社：Journal of Theoretical and Applied
摘要：Data integration is inevitable in current distributed heterogeneous database environment. The existing data integration system uses data federation to integrate the distributed data without duplicating the data sources or creating a new integrated data store for analysis and decision support. The contemporary data federation techniques convert the data sources to XML dataset and integrate the data. These systems do not resolve all conflicts efficiently. This paper proposes a layered Service Oriented Ontology based Data federation Architecture [SOOD] and hybrid ontology based data federation methodology over the SOOD architecture. This converts the metadata of data sources into OWL description for constructing global ontology using Local as View [LAV] approach for data federation. In the proposed method global ontology construction process uses the OWL description to resolve all schema level heterogeneities of data sources and guarantees quality data federation. The heterogeneities are classified in to two (schematic and data) and are resolved at two different stages of proposed ontology based data federation methodology. The storage infrastructure of SOOD architecture aids to store and fetch the results swiftly. Experimental results shows that the proposed methodology is relatively efficient than the conventional hybrid ontology based data federation methodology in recall and other quality parameters like response time and reusability. It is observed that the recall of SOOD system is increased by 10% and response time is reduced by 20% than the conventional hybrid ontology system.
关键词：Data Federation; Data Integration; Ontology; Resolving Heterogeneity; Service Oriented Computing
