Ontology-based model for resolving the data-level and semantic-level conflicts

Abstract

Mediating heterogeneous data sources through ontology-based data integration has been a critical issue in the database community for the past few years. To overcome the shortcoming of the traditional conflict resolutions for the ontology-based data integration system, this paper puts forward a new architecture called RCM, which provides an extensible mechanism in the descriptions of mapping formulas and constraints of conflicts between the global schema and the local schema. Grounded on the given model, an algorithm is proposed to deal with difficulties caused by data-level and semantic-level conflicts. It extracts formulas and constraints from mapping documents to analyze conflicts and then delivers them to corresponding handlers to modify the incorrect sub-queries. Thus, it identifies and resolves a variety of conflicts in heterogeneous databases so as to obtain the correct results of operations on the global schema. In addition, we present evaluation results to show that the model can be successfully used to implement the process of identifying and resolving conflicts.