Improved clustering technique for ITI-PrefixSpan

Abstract

Sequential data mining is the process to find out the frequent sub-sequences from the given sequential dataset. Sequential pattern mining can only reveal the sequence (order) of items, but it does not determined the time interval between two successive events. Time interval sequential mining is process to find out sequential patterns with time interval between two successive events. In this paper, we will introduce the new cluster technique so we will get dynamic cluster range rather than fixed. We improve the result of new ITI-PrefixSpan and compare our algorithm with other algorithms in terms of computing time and memory.