Skip to main content
. 2015 Jun 10;31(12):i320–i328. doi: 10.1093/bioinformatics/btv225

Fig. 6.

Fig. 6.

Evaluation of k-means clustering quality as a function of the number of clusters (average and standard deviation on 10 algorithm initializations). The same protocol was applied to a subset of the Mitocheck dataset and two samples of the same dimensions, respectively, drawn from the Uniform and the Normal distributions