Mining association rules using fast algorithm

Abstract

The most time consuming operation in Priori-like algorithms for association rule mining is the computation of the frequency of the occurrences of itemsets (called candidates) in the database. In this paper, a fast algorithm has been proposed for generating frequent itemsets without generating candidate itemsets and association rules with multiple consequents. The proposed algorithm uses Boolean vector with relational AND operation to discover frequent itemsets. Experimental results shows that combining Boolean Vector and relational AND operation results in quickly discovering of frequent itemsets and association rules as compared to general Apriori algorithm.