On the Sequential Pattern Mining Algorithm Based on Projection position

In order to avoid huge amount of projected databases produced by Prefix Span algorithm and reduce unnecessary storage space and scanning time, an improved sequential pattern mining algorithm based on projection position is proposed. The idea of improved algorithm is as follows: (1) utilizing Apriori property to delete the non-frequent items and divide search space, which can reduce unnecessary storage space and scanning time; (2) recording projected position to locate projected sequence position for mining local frequency items and mine each recursively so as to avoid physically constructing corresponding projected databases. The experimental results show that proposed algorithm has better feasibility and scalability compared with other algorithms.