An Improved Top-Down Data Mining Algorithm for Long Frequents

In this paper, in order to improve the method of computing support of candidate frequent itemsets, in order to reduce the times of scanning database when computing support, and in order to fast search long frequent itemsets, aiming to top-down search strategy, we propose an improved top-down association rules mining algorithm based on sequence number, which is suitable for mining long frequent itemsets since this top-down search strategy is adopted. The algorithm uses the way of binary Boolean calculation to generate binary candidate frequent itemsets, and uses the method of computing sequence number degree (SND) to obtain support of candidate frequent itemsets, which is gained through computing these sequence number (SN) of all these items in candidate frequent itemsets. The algorithm only need scan once database to indeed improve the efficiency of algorithm. The experiment indicates the efficiency of this algorithm is faster and more efficient than presented these similar algorithms when mining long frequent itemsets.