Modeling tools for the integration of structured data sources

Abstract

Disparity in representations within structured documents such as XML or SQL makes interoperability challenging, error-prone and expensive. A model is developed to process disparate representations to an encompassing generic knowledge representation. Data sources were characterized according to a number of smaller models: their case; the underlying data storage structures; a content model based on the ontological structure defined by the documents schema; and the data model or physical structure of the schema. In order to harmonize different representations and give them semantic meaning, from the above categories the representation is mapped to a common dictionary. The models were implemented as a structured data analysis tool and a basis was built to compare across schema and documents. Data exchange within modeling and simulation environments are increasingly in the form of XML using a variety of schema. Therefore, we demonstrate the use of this modeling tool to automatically harmonized multiple disparate XML data sources in a prototype simulated environment.