An array based approach for mining maximal frequent itemsets

Abstract

Mining of frequent patterns is a basic problem in data mining applications. The algorithms which are used to generate the frequent patterns must perform efficiently. The objective was to propose a new algorithm which generates maximal frequent patterns in less time. We proposed an algorithm which was based on Array technique and combines a vertical tidset representation of the database with effective pruning mechanisms. It removes all the nonmaximal frequent itemsets to get exact set of MFI directly. It works efficiently when the number of itemsets and tidsets are more. The proposed approach has been compared with GenMax algorithm for mushroom dataset and the results show the proposed algorithm generates less number of candidate itemsets to find all MFIs. Hence, the proposed algorithm performs effectively and generates frequent patterns faster.