Closed Sequential Pattern Mining in High Dimensional Sequences

High dimensional sequences, such as biological sequences, are characterized by a small number of transactions, and a large number of items in each transaction. Mining sequential patterns in the sequences need to consider different forms of patterns, such as contiguous patterns, local patterns which appear more than one time in a special sequence, and so on. Mining closed patterns might lead to not only a more compact complete result set, but also better efficiency. In this paper, a novel algorithm based on BIDE (BI-Directional Extension) and multi-support is presented for high dimensional sequences specifically. It mainly mines three types of closed sequential patterns which are sequential patterns, local sequential patterns and total sequential patterns. Thorough experimental performances on biological sequences have demonstrated that the proposed algorithm could reduce memory consumption and generate more compact patterns.