Résumé: In this article, we introduce pseudoconstraints, a novel data mining pattern aimed at identifying rare events in databases. At first, we formally define pseudoconstraints using a probabilistic model and provide a statistical test to identify pseudoconstraints in a database. Then, we focus on a specific class of pseudoconstraints, named cycle pseudoconstraints, which often occur in databases. We define cycle pseudoconstraints in the context of the ER model and present an automatic method for detecting cycle pseudoconstraints from a relational database. Finally, we present an experiment to show cycle pseudoconstraints “at work” on real data.

Langue: Anglais