Efficient modification of fast updated FP-trees based on pre-large concepts

In this paper, we propose a new FUFP-tree maintenance algorithm, based on the concept of pre-large itemsets, in order to efficiently handle record modification. Pre-large itemsets are defined by a lower support threshold and an upper support threshold. This helps to reduce rescans of the original database. The proposed approach can get good execution time for tree maintenance, especially when a small number of records are modified. Experimental results showed that the proposed Pre-FUFP modification algorithm performed well when handling updated records and generated nearly the same tree structure as the original FP tree algorithm.