The paper “Mining Constraint Violations” by Stefano Ceri, Francesco Di Giunta, and myself has been accepted for publication on the ACM Transactions on Database Systems.
Abstract. In this paper, we introduce pseudo–constraints, a novel data mining pattern aimed at identifying rare events in databases. At first, we formally define pseudo–constraints using a probabilistic model and provide a statistical test to identify pseudo–constraints in a database. Then, we focus on a specific class of \pcs, named cycle pseudo–constraints, which often occur in databases. We define cycle pseudo–constraints in the context of the ER model and present an automatic method for detecting cycle pseudo–constraints from a relational database. Finally we present an experiment to show cycle pseudo–constraints “at work” on real data.
