Skip to main content

Table 1 The average NMIs/ACCs and the standard errors obtained by the ECMC and other comparison partners in seven simulation data sets

From: Subtype identification from heterogeneous TCGA datasets on a genomic scale by multi-view clustering with enhanced consensus

  Methods Simulation 1 Simulation 2
   t = 0.95 t = 1 t = 1.2 t =2 s = 25 s= 30 s =40 s = 50
NMI SV1 0.856 0.465 0.007 0.006 0.524 0.470 0.404 0.337
  SV2 0.775 0.495 0.012 0.002 0.524 0.470 0.421 0.331
  Concat 0.919 0.696 0.021 0.006 0.527 0.472 0.421 0.340
  Coreg 0.919 0.566 0.344 0.007 0.542 0.491 0.421 0.344
  SNF 0.960 0.889 0.012 0.005 0.562 0.510 0.421 0.519
  CMC 0.919 0.542 0.493 0.335 0.594 0.503 0.480 0.744
  ECMC 1.000 0.882 1.000 1.000 0.667 1.000 0.859 1.000
ACC SV1 0.975 0.878 0.550 0.545 0.875 0.850 0.800 0.745
  SV2 0.960 0.886 0.565 0.525 0.875 0.850 0.800 0.750
  Concat 0.990 0.945 0.585 0.545 0.875 0.850 0.800 0.748
  Coreg 0.990 0.910 0.770 0.550 0.875 0.850 0.800 0.750
  SNF 0.995 0.985 0.565 0.540 0.875 0.850 0.800 0.780
  CMC 0.990 0.890 0.777 0.701 0.875 0.850 0.800 0.890
  ECMC 1.000 0.975 1.000 1.000 0.898 1.000 0.980 1.000
  1. Highest NMIs/ACCs are marked in bold