Extraction of Non-redundant Association Rules from Concept Lattices based on IsoFCA System

The problem of association rule mining is one of the most frequently studied and popular KDD tasks. Association rule mining is an important sub-branch of data mining. Based on the known current existence of association rule mining algorithms, this paper emphasizes the research work of deleting redundant association rules. It is a problem to mine quantitative association rules because the existing algorithms always generate too many association rules. Through analyzing the mathematics properties of concept lattice and characteristics of redundant association rules, we design an algorithm which can get the minimal antecedents and maximal consequents sets of rules. These sets can be built by the intent relations between a sub-concept and its sup-concepts as well as the minimal set nature of intent of concept. Using this algorithm we can deduce other rules with the same minimal support and minimal confidence and get all frequent class association rules that satisfy the minimum support. Finally, we integrate this algorithm into an IsoFCA system and it has been successfully applied in practice.