Finding Exception For Association Rules Via SQL Queries

Abstract

Finding association rules is mainly based on generating larger and larger frequent set candidates, starting from frequent attributes in the database. The frequent sets can be organised as a part of a lattice of concepts according to the Formal Concept Analysis approach. Since the lattice construction is database contents-dependent, the pseudo-intents (see Formal Concept Analysis) are avoided. Association rules between concept intents (closed sets) A=>B are partial implication rules, meaning that there is some data supporting A and (not B); fully explaining the data requires finding exceptions for the association rules. The approach applies to Oracle databases, via SQL queries.