Summary: Lehrstuhl für Informatik 9
Zeit: Dienstag, 2. März 2010, 15.00 Uhr
Ort: Seminarraum 5056, Ahornstr. 55
Referent: Dr. Arthur Zimek
Titel: Clustering in Subspaces of High-Dimensional Data
There are a wide range of different solutions to the problem of clustering in sub-
spaces of high-dimensional data, but even the problem is not always clearly defined.
This talk sketches the general problem, investigates different specialized formulations
of the problem, and identifies accordingly different families of solutions like subspace
clustering, projected clustering, bi-clustering (co-clustering), correlation clustering
(arbitrarily oriented subspace clustering). Finally, an approach to correlation cluster-
ing proposed by the author is discussed in more detail.
Arthur Zimek is a post doc in the database systems and data mining group of Hans-
Peter Kriegel at the Ludwig-Maximilians-Universität München, Germany. He holds
degrees in bioinformatics, philosophy, and theology, involving studies at universities