Implementation of Sequential Pattern Mining with Item Intervals

Abstract

Sequential Pattern mining is an important data mining field with wide range of applications that can extract frequent sequences while maintaining their order. It is important to identify item intervals of sequential patterns extracted by sequential pattern mining. There are two approaches for integration of item intervals with sequential pattern mining; constraint-based mining and extended sequence-based mining. This paper presents the combination of those two item interval approaches. PrefixSpan algorithm is used to find the frequent sequence patterns from the sequence database. PrefixSpan algorithm overcomes the problems of Apriori-based algorithms since it avoids the candidate generation and multiple database scanning time. Moreover, prefix-projectiong substantially reducest the size of projected databases and leads to efficient processing.