Towards a Context Ontology to Enhance Data Integration Processes

Data Integration has become one of the most relevant research fields in data management for the last years. The problem of integrating data from diverse, heterogeneous data sources is ubiquitous and has been tackled by some different approaches. A recent one concerns Peer Data Management Systems (PDMS) which are characterized by their dynamicity. To help matters, semantic information like context may be employed to ease some processes in DI: schema reconciling and query processing. However, dealing with contextual information entails a high development cost because several tasks (e.g. context acquisition and processing) must be performed. In order to provide means for that, first we have to define how to better represent contextual information. In that respect, ontologies are an interesting approach since they enable reasoning, reusability and knowledge sharing. In this paper, we propose CODI a Context Ontology to formally represent context in Data Integration processes. We also present a case study illustrating how CODI can be used to enhance query processing in a PDMS environment, so that users will get more complete and relevant results.