If you are experiencing problems downloading PDF or HTML fulltext, our helpdesk recommend clearing your browser cache and trying again. If you need help in clearing your cache, please click here . Still need help? Email help@ingentaconnect.com

Clusterings should not be compared by visual inspection: response to Gagné & Proulx

$48.00 plus tax (Refund Policy)

Download / Buy Article:

Abstract:

Abstract

In Heikinheimo et al. (Journal of Biogeography, 2007, 34, 1053–1064) we used clustering to analyse European land mammal fauna. Gagné & Proulx criticized our choice of the Euclidean distance measure in the analysis, and advocated the use of the Hellinger distance measure, claiming that this leads to very different clustering results. The criticism fails to take into account the probabilistic nature of the methods used and the fact that in this case the similarity measures correlate strongly. Gagné & Proulx used subjective inspection as the criterion of similarity between clusterings. We show that this is insufficient and misleading. Namely, owing to the local minimum problem, two clustering runs rarely give identical results. In the case of our study, the measured similarity (using the kappa statistic) between the Euclidean- and Hellinger-based clusterings is roughly equal to the similarity between two clusterings that both use the Hellinger distance but different random initialization points.

Keywords: Clustering; Euclidean distance; Europe; Hellinger distance; double-zero problem; kappa; objective comparison; optimization with randomization; species abundance paradox

Document Type: Correspondence

DOI: http://dx.doi.org/10.1111/j.1365-2699.2008.02054.x

Affiliations: Department of Geology and Institute of Biotechnology, FIN-00014 University of Helsinki, PO Box 64, Helsinki, Finland

Publication date: March 1, 2009

Related content

Tools

Favourites

Share Content

Access Key

Free Content
Free content
New Content
New content
Open Access Content
Open access content
Subscribed Content
Subscribed content
Free Trial Content
Free trial content
Cookie Policy
X
Cookie Policy
ingentaconnect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more