Algorithm for sequence mining using gap constraints

The sequence mining problem consists in finding frequent sequential patterns in a database of timestamped events. Some application domains require limiting the maximum temporal gap between events in the input sequences. However concentration on such constraint is critical for most sequence mining algorithms. In this paper we describe CCSM (Cache-based Con-strained Sequence Miner), a new level-wise algorithm that overcomes the troubles usually related to this kind of con-straints. CCSM adopts an approach based on k-way intersections of idlists to compute the support of candidate sequences. Our k-way intersection method is enhanced by the use of an effective cache that stores intermediate idlists for reuse in future. The reuse of intermediate results involves a surprising reduction in the actual number of join operations performed on idlists. CCSM has been experimentally compared with cSPADE, a state of the art algorithm, on several synthetically generated datasets, achieve better or similar results in most cases.