Authors:
Luciano Caroprese
;
Sergio Greco
and
Ester Zumpano
Affiliation:
DEIS, Università della Calabria, Italy
Keyword(s):
Data Integration, Inconsistent Databases, Consistent Query Answers.
Related
Ontology
Subjects/Areas/Topics:
Coupling and Integrating Heterogeneous Data Sources
;
Databases and Information Systems Integration
;
Enterprise Information Systems
Abstract:
In this paper a framework for merging, repairing and querying inconsistent databases is presented. The framework, considers integrity constraints defining primary keys, foreign keys and general functional dependencies. The approach consists of three steps: i) merge of the source databases by means of integration operators or general SQL queries, to reduce the set of tuples coming from the source databases which are inconsistent with respect to the constraints defined by the primary keys, ii) repair of the integrated database by completing and/or cleaning the set of tuples which are inconsistent with respect to the inclusion dependencies (e.g. foreign keys), and iii) compute consistent answers over repaired databases which could be still inconsistent with respect to the functional dependencies. The complexity of merging, repairing and computing consistent answers will be show to be polynomial and a prototype of a system integrating databases and computing queries over possible inconsi
stent databases will be presented.
(More)