Re-Filterization of Negative Association Rules Based On Rule of Importance

Abstract

Genarally positive frequent itemsets are considered for generation of association rules. Which indicates the presence of item in a transactional Database. In this paper, the complementary problem i.e. negative association rule mining is also considered along with positive association rule mining. The proposed method utilizes Item based Bit Pattern and strong pruning mechanism at Item level as well as rule level. Interest based measures are added to the proposed method to reduce the positive and negative rules which are not interested. Refilterization mechanism is used to extract meaningful negative association rules according to the market demands