A Survey on Mining User-Aware Rare Sequential Pattern

Salmath Amina KP, Farzin Ahammed T

Abstract: Sequential pattern mining is exploring enthusiasm for finding the most frequently using patterns from the data sets. We explore the sequential pattern mining using a bitmap representation. Our algorithm incrementally output new frequent itemset in an online fashion using depth-first search strategy. To find out the sequential pattern in data streams, we propose a sequential data mining algorithm using weighted sliding window. SWSS algorithm provides more space for users to specify which sequences they are more interested in. To address the problem of finding frequent and rare itemsets from the data streams, we propose another algorithm called NewSPAM. NewSPAM Algorithm is the combination of both PerSPAM and MinSPAM algorithm.

Keywords: Frequent Itemset Mining, SWSS, Rare itemset, Rare Sequential pattern mining, Weighted Sliding Window