A New Quartet Tree Heuristic for Hierarchical Clustering
Rudi Cilibrasi and Paul Vitányi
unpublished yet (in progress)
We present a new quartet tree heuristic for
from weighted quartet topologies, and a standard manner
to derive those from a given distance matrix.
We do not assume that there is a true ternary tree that generated the
quartet topologies or distances which
we wish to recover as closely as possible.
Our aim is to just model the input data as faithfully as possible
by the quartet tree. Our method is capable of handling up to 60--80
objects in a matter of hours, while no existing quartet heuristic
can directly compute a quartet tree of more than about 20--30 objects
without running for years.
The method is implemented and available as public software.