Computation-time efficient and robust attribute tree mining with DryadeParent
A Termier, M-C Rousset, Michele Sebag, K Ohara, T Washio and Hiroshi Motoda
In: ECML/PKDD 2005, 03-07 Oct 2005, Porto, Portugal.
In this paper, we present a new tree mining algorithm, Dryade-
Parent, based on the hooking principle first introduced in Dryade .
In the experiments, we demonstrate that the branching factor and depth
of the frequent patterns to find are key factor of complexity for tree min-
ing algorithms, even if often overlooked in previous work. We show that
DryadeParent outperforms the current fastest algorithm, CMTreeM-
iner, by orders of magnitude on datasets where the frequent patterns
have a high branching factor.