An efficient algorithm for mining sequential generator pattern using prefix trees and hash tables

摘要: Mining long frequent sequences that contain a combinatorial number of subsequences or using very low support thresholds to mine sequential patterns is both time-and memory-consuming. The mining closed patterns, generator and maximum has been proposed overcome this problem. This paper proposes an algorithm for generating all patterns. uses vertical approach listing counting the sequence based on prime block encoding represent candidate determine frequency each candidate. search space much smaller than those other algorithms because super frequency-based pruning non-generator-based are applied. Besides, hash tables also used fast checking existed Experimental results conducted synthetic real databases show effective.