DryadeParent, An Efficient and Robust Closed Attribute Tree Mining Algorithm

Alexandre Termier , Marie-Christine Rousset (1) , Michèle Sebag (2) , Kouzou Ohara , Takashi Washio , Hiroshi Motoda

1 IMAG - Institut d'Informatique et de Mathématiques Appliquées de Grenoble
2 LRI - Laboratoire de Recherche en Informatique

In this paper we present a new tree mining algorithm, DryadeParent, 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 factors of complexity for tree mining algorithm, even if often overlooked in previous work. We show that DryadeParent outperforms the current fastest algorithm, CMTreeMiner, by orders of magnitude on datasets where the frequent tree patterns have a high branching factor.

Domaines: Intelligence artificielle [cs.AI]

Alexandre Termier, Marie-Christine Rousset, Michèle Sebag, Kouzou Ohara, Takashi Washio, et al.. DryadeParent, An Efficient and Robust Closed Attribute Tree Mining Algorithm. IEEE Transactions on Software Engineering, 2008. ⟨inria-00292914⟩