Comprehensive study of weighted Sequential Pattern Mining

Extensive growth of data gives the motivation to find meaningful patterns among the huge data. Sequential pattern provides us interesting relationships between different items in sequential database. In the real world, there are several applications in which specific sequences are more important than other sequences. Traditional Sequential pattern approaches are suffering from two disadvantages: Firstly, all the items and sequences are treated uniformly. Second, conventional algorithms are generating large number of patterns for lower support. In addition, the unimportant patterns with low weights can be detected. This paper addresses problem of traditional framework and various framework of weighted sequential pattern. Paper also discuses how algorithm mines sequential pattern which reduces the search space and new pruning technique prune the unimportant pattern and pick only those patterns which leads to important and emerging pattern. Later section of paper discuses results of simulation study and how researcher can lead current research.