Mining association rules using formal concept analysis

Abstract : In this paper, we give an overview of the use of Formal Concept Analysis in the framework of association rule extraction. Using frequent closed itemsets and their generators, that are defined using the Galois closure operator, we address two major problems: response times of association rule extraction and the relevance and usefulness of discovered association rules. We quickly review the Close and the A-Close algorithms for extracting frequent closed itemsets using their generators that reduce response times of the extraction, specially in the case of correlated data. We also present definitions of the generic and informative bases for association rules which generation improves the relevance and usefulness of discovered association rules.