An Argumentation-based Approach to Database Repair

The access to high-quality data is essential to make accurate decisions. Consequently, when a database becomes inconsistent is crucial to restore its consistency. The main approach for database consistency restoration is based on the notion of repair. In this work, we use Argumentation techniques to provide a better understand of the reasoning process behind database reparation. We introduced an extended argumentation framework that provides a comprehensive and alternative way to identify, represent and resolve the conflicts between tuples in an inconsistent database. We studied the complexity of this framework and show that it can be used to check the optimality of a repair based on the extended notions of locally, semi-globally and globally optimal repair with respect to Denial Constraints and Tuple-Generating Dependencies classes.