Skip to main content

Table 2 Test results for high dimensional synthetic data sets.

From: MULTI-K: accurate classification of microarray subtypes using ensemble k-means clustering

Data set

H2

H3

MULTI-K (Cut & Entropy)

0.9210 (9/10)

0.8695 (7/10)

GCCk

0.5367 (5/10)

0.7456 (8/10)

GCCc

0.0993 (0/10)

0.7799 (10/10)

Hier.

gap

0 (1/10)

0.2277 (1/10)

 

Silhuette

0.0839 (9/10)

0.3444 (4/10)

k -means

gap

0.7138 (5/10)

0.8433 (7/10)

 

Silhuette

0.8445 (9/10)

0.3715 (1/10)

  1. The adjusted Rand indexes are shown averaged over ten randomly generated data sets. In each parenthesis is shown the number of cases out of ten that correctly identified the number of clusters. ARI values that exceed 0.7 are shown in bold.