Back
Clustering: Science or Art?
We examine whether the quality of di erent clustering algorithms can be compared by a general, scienti cally sound procedure which is independent of particular clustering algorithms. We argue that the major obstacle is the diculty in evaluating a clustering algorithm without taking into account the context: why does the user cluster his data in the rst place, and what does he want to do with the clustering afterwards? We argue that clustering should not be treated as an application-independent mathematical problem, but should always be studied in the context of its end-use. Di erent techniques to evaluate clustering algorithms have to be developed for di erent uses of clustering. To simplify this procedure we argue that it will be useful to build a \taxonomy of clustering problems" to identify clustering applications which can be treated in a uni ed way and that such an e ort will be more fruitful than attempting the impossible | developing \optimal" domain-independent clustering algorithms or even classifying clustering algorithms in terms of how they work.
@inproceedings{6331, title = {Clustering: Science or Art?}, booktitle = {JMLR Workshop and Conference Proceedings, Volume 27}, abstract = {We examine whether the quality of di erent clustering algorithms can be compared by a general, scienti cally sound procedure which is independent of particular clustering algorithms. We argue that the major obstacle is the diculty in evaluating a clustering algorithm without taking into account the context: why does the user cluster his data in the rst place, and what does he want to do with the clustering afterwards? We argue that clustering should not be treated as an application-independent mathematical problem, but should always be studied in the context of its end-use. Di erent techniques to evaluate clustering algorithms have to be developed for di erent uses of clustering. To simplify this procedure we argue that it will be useful to build a \taxonomy of clustering problems" to identify clustering applications which can be treated in a uni ed way and that such an e ort will be more fruitful than attempting the impossible | developing \optimal" domain-independent clustering algorithms or even classifying clustering algorithms in terms of how they work.}, pages = {65-79}, organization = {Max-Planck-Gesellschaft}, school = {Biologische Kybernetik}, year = {2012}, slug = {6331}, author = {von Luxburg, U. and Williamson, R. and Guyon, I.} }