Author(s)

Feng Fan, Liao Husheng, Jin Xueyun

Corresponding Author

Fan Feng

ABSTRACT

Frequent sequential pattern mining is an important field in data mining. Compared with the static data, the stream data is a single scan data obtained in a continuous and real-time way. The frequent pattern mining algorithm of traditional static sequence database has been difficult to meet the frequent pattern mining requirements for streaming data. The traditional serial processing method is time-consuming and cannot meet the requirements of high performance processing. Based on the existing Pisa algorithm, this paper presents a parallel algorithm named Parallel-Pisa, it can adjust the parallel strategy according to the different velocity of the stream data to improve the efficiency of the algorithm so that it can be better applied to frequent sequence pattern mining of stream data.

KEYWORDS

Frequent sequential pattern mining, Stream data, Parallel processing, Self-adaption.