Dynamic load balancing algorithms for sequence mining

Mining sequential patterns in large database is an important problem in data mining research. Enormous sizes of available datasets and possibly large number of mined patterns demand efficient and scalable algorithms. In this paper, we present a new dynamic load algorithm based HPSPM (hash-based parallel algorithm for mining sequential patterns) on shared-nothing environment. Experiments on Dawning 300 cluster system show that this algorithm achieves good speedup and is substantially improved compared to HPSPM.