Towards a standard methodology to evaluate internal cluster validity indices

► Internal cluster validity indices (CVIs) are used to evaluate cluster partitions. ► The usual CVI comparison methodology makes an assumption which is often incorrect. ► We propose an alternative methodology to compare CVIs that obtains better results. The evaluation and comparison of internal clus...

Full description

Saved in:
Bibliographic Details
Published in:Pattern recognition letters 2011-02, Vol.32 (3), p.505-515
Main Authors: Gurrutxaga, Ibai, Muguerza, Javier, Arbelaitz, Olatz, Pérez, Jesús M., Martín, José I.
Format: Article
Language:English
Subjects:
Applied sciences
Cluster validation
Cluster validity index
Exact sciences and technology
Information, signal and communications theory
Signal and communications theory
Signal representation. Spectral analysis
Signal, noise
Telecommunications and information theory
Unsupervised learning
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:► Internal cluster validity indices (CVIs) are used to evaluate cluster partitions. ► The usual CVI comparison methodology makes an assumption which is often incorrect. ► We propose an alternative methodology to compare CVIs that obtains better results. The evaluation and comparison of internal cluster validity indices is a critical problem in the clustering area. The methodology used in most of the evaluations assumes that the clustering algorithms work correctly. We propose an alternative methodology that does not make this often false assumption. We compared 7 internal cluster validity indices with both methodologies and concluded that the results obtained with the proposed methodology are more representative of the actual capabilities of the compared indices.
ISSN:0167-8655
1872-7344
DOI:10.1016/j.patrec.2010.11.006