Combining Clustering techniques and FCA to characterize Interestingness Measures

Formal Concept Analysis "FCA" is a data analysis method which enables to discover hidden knowledge existing in data. A kind of hidden knowledge extracted from data is association rules. Di erent Interestingness Measures "IMs" were reported in the literature to extract only relevant association rules. Given a dataset, the choice of a good interestingness measure remains a challenging task for a user. Given an interestingness measures evaluation matrix according to semantic properties, this paper describes how FCA can highlight interestingness measures with similar behavior in order to help the user during his choice. The aim of this paper is the discovery of interestingness measures clusters, able to validate those found due to a method of agglomerative hierarchical clustering (AHC) using Ward criterion and a method of a non-hierarchical clustering of k-means. Then, based on the theoretical study of sixty one interestingness measures according to nineteen properties, proposed in a recent study, FCA describes several groups of measures.