Skip to main content

Table 4 Performance of the classifiers on real gene expression data sets for the two class classification tasks

From: Improved shrunken centroid classifiers for high-dimensional class-imbalanced data

Data set Method λ a # genes PA PA1 PA2 g-means AUC
Ivshina PAM 0 22283 0.84 0.79 0.85 0.82 0.85
(ER) GM-PAM 4.83 51 0.82 0.91 0.81 0.86 0.90
  ALP 0 22283 0.84 0.82 0.84 0.83 0.86
  GM-ALP 58.24 26 0.85 0.91 0.84 0.87 0.88
  AHP 185.56 20 0.89 0.88 0.89 0.88 0.90
  GM-AHP 69.58 115 0.89 0.88 0.89 0.89 0.91
Wang PAM 3.71 14 0.61 0.60 0.62 0.61 0.62
  GM-PAM 3.71 14 0.60 0.60 0.62 0.61 0.63
  ALP 8.26 654 0.56 0.57 0.55 0.56 0.63
  GM-ALP 8.26 654 0.56 0.56 0.56 0.56 0.63
  AHP 21.95 135 0.56 0.56 0.56 0.56 0.65
  GM-AHP 21.95 135 0.56 0.56 0.55 0.56 0.63
Korkola PAM 0.19 7073 0.65 0.71 0.57 0.64 0.64
  GM-PAM 0.19 7073 0.65 0.71 0.57 0.64 0.64
  ALP 4.87 155 0.64 0.65 0.62 0.63 0.64
  GM-ALP 4.87 155 0.69 0.74 0.62 0.67 0.69
  AHP 0.76 1308 0.58 0.68 0.43 0.54 0.58
  GM-AHP 0.76 1308 0.62 0.71 0.48 0.58 0.60
  1. The table reports the same information as Table 1; # genes is the number of active genes. Optimal thresholds were estimated with 5-fold CV and the accuracy measures with LOOCV; see text for details.
  2. aFor AHP and GM-AHP only λ θ was optimized while λ γ was set to zero.