Clustering as an Example of Optimizing Arbitrarily Chosen Objective Functions

This source preferred by Marcin Budka

Authors: Budka, M.

Editors: Nguyen, N., Trawinski, B., Katarzyniak, R. and Geun-Sik, J.

http://eprints.bournemouth.ac.uk/20472/

Volume: 457

Pages: 177-186

Publisher: Springer Berlin / Heidelberg

ISBN: 978-3-642-34299-8

DOI: 10.1007/978-3-642-34300-1_17

This paper is a reflection upon a common practice of solving various types of learning problems by optimizing arbitrarily chosen criteria in the hope that they are well correlated with the criterion actually used for assessment of the results. This issue has been investigated using clustering as an example, hence a unified view of clustering as an optimization problem is first proposed, stemming from the belief that typical design choices in clustering, like the number of clusters or similarity measure can be, and often are suboptimal, also from the point of view of clustering quality measures later used for algorithm comparison and ranking. In order to illustrate our point we propose a generalized clustering framework and provide a proof-of-concept using standard benchmark datasets and two popular clustering methods for comparison.

This data was imported from DBLP:

Authors: Budka, M.

Editors: Nguyen, N.T., Trawinski, B., Katarzyniak, R. and Jo, G.

http://eprints.bournemouth.ac.uk/20472/

http://dx.doi.org/10.1007/978-3-642-34300-1

Volume: 457

Pages: 177-186

Publisher: Springer

ISBN: 978-3-642-34299-8

DOI: 10.1007/978-3-642-34300-1_17

This data was imported from Scopus:

Authors: Budka, M.

http://eprints.bournemouth.ac.uk/20472/

Volume: 457

Pages: 177-186

ISBN: 9783642342998

DOI: 10.1007/978-3-642-34300-1-17

This paper is a reflection upon a common practice of solving various types of learning problems by optimizing arbitrarily chosen criteria in the hope that they are well correlated with the criterion actually used for assessment of the results. This issue has been investigated using clustering as an example, hence a unified view of clustering as an optimization problem is first proposed, stemming from the belief that typical design choices in clustering, like the number of clusters or similarity measure can be, and often are suboptimal, also from the point of view of clustering quality measures later used for algorithm comparison and ranking. In order to illustrate our point we propose a generalized clustering framework and provide a proof-of-concept using standard benchmark datasets and two popular clustering methods for comparison. © Springer-Verlag Berlin Heidelberg 2013.

This data was imported from Web of Science (Lite):

Authors: Budka, M.

http://eprints.bournemouth.ac.uk/20472/

Volume: 457

Pages: 177-186

ISBN: 978-3-642-34299-8

DOI: 10.1007/978-3-642-34300-1_17

The data on this page was last updated at 04:42 on September 20, 2017.