A New Association Rules Mining Algorithm Based on Vector

Abstract

As a classical algorithm of association rules mining, Apriori algorithm has two bottlenecks: the large number of candidate itemsets and the poor efficiency of counting support. A new association rules mining algorithm based on vector is proposed, which can reduce the number of candidate frequent itemsets, improve efficiency of pruning operation and count support quickly using vector inner product operation and vector addition operation between transaction vector and itemset vector. According to the results of the experiments, the proposed algorithm can quickly discover frequent itemsets and is more efficient than Apriori algorithm.