A Skipping FP-Tree for Incrementally Intersecting Closed Itemsets in On-Line Stream Mining
Takumi Nishina, Koji Iwanuma, Yoshitaka Yamamoto · 2019
An on-line mining for a data stream consisting of large transactions is still quite difficult because of an explosion of frequent itemsets. In this paper, we propose a new data structure, called a skipping FP-tree, which enables us to effectively compress the set of closed itemsets in a stream. We show, through experimental evaluations, the skipping FP-tree achieves more than ten times faster computation for incremental intersections.