A block-based approach for frequent itemset mining over data streams
Mina Memar, Mohammad Hadi Sadreddini, Mahmood Deypir, Seyed Mostafa Fakhrahmad · 2011
Sliding window is a widely used model for data stream processing and mining. Frequent itemset mining over sliding window is a challenging problem due to limited processing resources. In this study, an efficient representation of sliding window is proposed. In this representation, a blocked bit sequence technique is used to enhance both sliding and mining time. Experimental evaluations show that our algorithm outperforms a recently proposed algorithm.