We here introduce the notions which are fundamental for dealing with the problem of extracting reliable information from (inconsistent) data in the presence of aggregate constraints. Specifically, we formalize the basic concepts of aggregate constraint, repair, repair minimality, and consistent answer, which will be exploited in the following chapters to define a general framework for repairing and querying inconsistent numerical data. We also provide a brief overview of the complexity classes which will be referred to when addressing the complexity characterization of several problems related to the above-mentioned framework.
Unable to display preview. Download preview PDF.