Cluster quality measures

Question

Does Matlab provide any facility for evaluating clustering methods? (cluster compactness and cluster separation. ....) Or is there any toolbox for it?

score 5 · Answer 1 · answered May 10 '13 at 12:00

Matlab provides Silhouette index and there is a toolbox CVAP: Cluster Validity Analysis Platform for Matlab. Which includes following validity indexes:

Davies-Bouldin
Calinski-Harabasz
Dunn index
R-squared index
Hubert-Levin (C-index)
Krzanowski-Lai index
Hartigan index
Root-mean-square standard deviation (RMSSTD) index
Semi-partial R-squared (SPR) index
Distance between two clusters (CD) index
weighted inter-intra index
Homogeneity index
Separation index

Note that you might need precompiled LIBRA binaries for your platform.

score 4 · Answer 2 · answered Apr 28 '12 at 19:57

4

You can try Silhouette plot from the Statistical toolbox.

For an example see this documentation.

answered Apr 28 '12 at 19:57

yuk

19,098
13
68
99

Thank. But I want use validation function as fitness/goal function for particle swarm clustering. – Handsome Nerd Apr 28 '12 at 20:10

score 4 · Accepted Answer · answered Apr 28 '12 at 21:21

4

Not in Matlab, but ELKI (Java) provides a dozen or so cluster quality measures for evaluation.

answered Apr 28 '12 at 21:21

Has QUIT--Anony-Mousse

76,138
12
138
194

score 3 · Answer 4 · answered Sep 26 '13 at 19:04

Be aware that the Silhouette in Matlab has some strange behavior for singleton clusters. It assigns a score of 1 for singletons, when, for me, a more reasonable approach would be to give 0 for these clusters. In the Matlab implementation, if you give number of clusters as number of objects, Silhouette will give you a score of 1.

Cluster quality measures

4 Answers4

Linked