The "webkbn" dataset
From Thomas Finley, tomf@cornell.edu

USE

This is a dataset for supervised clustering.  See
http://www.cs.cornell.edu/~tomf/projects/supervisedkmeans
for more information.

In this directory there are data files of the form

	{cornell,texas,washington,wisconsin}.svmdata

representing a single clustering of the data, with each clustering coming from a single university.

The "cluster" files are collations of multiple clusterings, and are fairly straightforward.  For example,

	cluster.co
	cluster.tewi
	cluster.notwa

Are the set of clusterings of the single element set of the "cornell.svmdata" clustering, the "texas.svmdata" and "wisconsin.svmdata" clusterings, and all the clusterings *except* for the "washington" clustering.

LICENSING TERMS 

This dataset is granted free of charge for research and education
purposes. However you must obtain a license from the author to use it
for commercial purposes.  The terms of LICENSE.txt included in this
archive also apply.

Scientific results produced using the data provided shall cite this
as the source paper

       T. Finley, T. Joachims, Supervised k-means Clustering,
       SIGKDD, 2008.
