Discovery of high utility rare itemsets using PCR tree

Abstract

Data mining is used to extract interesting relationships between data in a large database. High utility rare itemsets in a transaction database can be used by retail stores to adapt their marketing strategies in order to increase their profits. Even though the itemsets mined are infrequent, since they generate a high profit for the store, marketing strategies can be used to increase the sales of these items. In this paper, a new method called the PCR tree method is proposed to generate all high utility rare itemsets while keeping the algorithm time-efficient. The proposed method generates the itemsets in one scan of the database. Results show that the time taken by the proposed method is nearly half that of the existing method, i.e. the UPR tree.