An Efficient ,Robust Closed Attribute Tree Mining Algorithm

     In my article, I present a new tree mining              In my article, I focus on
algorithm, DRYADEPARENT, based on the hookingtree mining that is, finding frequent tree-shaped
principle first introduced in DRYADE. In thepatterns in a database of tree-shaped data. Tree
project, I demonstrate that the branching factormining can lead to many practical applications in
and depth of the frequent patterns to find arethe areas of computer networks, bioinformatics,
key factors of complexity for tree miningand XML documents databases mining and hence
algorithms, even if often overlooked in previoushave received a lot of attention from the
algorithm. I show that DRYADEPARENTresearch community in recent years. Most of the
outperforms the current fastest algorithm,well-known algorithms use the same
CMTreeMiner, by orders of magnitude on datagenerate-and-test principle that made the success
sets where the frequent tree patterns have aof frequent item set algorithms. The main
high branching factor. DRYADE is based on aadaptation to the tree case is the design of
more general tree inclusion definition appropriateefficient candidate tree enumeration algorithms in
for mining highly heterogeneous collections of treeorder to avoid generating redundant candidates
data. DRYADEPARENT follows the same principlesand to enable efficient pruning. However, the
of DRYADE but uses a standard inclusionsearch space of tree candidates is huge,
definition. The performance of theparticularly when the frequent trees to find have
DRYADEPARENT is very fast when compare toboth a high depth and a high branching factor.
the CMTreeMiner.