Skip to main content

Table 2 Classification performances

From: Building gene expression profile classifiers with a simple and efficient rejection option in R

  G 3 G 4 G 5 G 6
Classifier Sens CI Spec CI Sens CI Spec CI Sens CI Spec CI Sens CI Spec CI
K-NN + rule 0.70 [0.61-0.79] 0.42 [0.31-0.53] 0.70 [0.65-0.75] 0.45 [0.36-0.55] 0.59 [0.49-0.69] 0.42 [0.35-0.49] 0.55 [0.49-0.62] 0.59 [0.48-0.70]
RF + rule 0.61 [0.53-0.69] 0.76 [0.69-0.83] 0.75 [0.71-0.79] 0.68 [0.60-0.77] 0.69 [0.66-0.73] 0.78 [0.70-0.86] 0.77 [0.70-0.84] 0.63 [0.53-0.73]
N-NET+rule 0.38 [0.23-0.53] 0.68 [0.53-0.83] 0.80 [0.70-0.90] 0.50 [0.43-0.57] 0.57 [0.51-0.63] 0.71 [0.59-0.84] 0.81 [0.75-0.87] 0.34 [0.25-0.44]
K-NN-OC SOCMT 0.74 [0.67-0.82] 0.20 [0.17-0.23] 0.90 [0.87-0.93] 0.29 [0.23-0.35] 0.91 [0.89-0.92] 0.23 [0.21-0.25] 0.86 [0.81-0.90] 0.16 [0.10-0.21]
KMeans-OG SOCM 0.47 [0.41-0.53] 0.66 [0.60-0.73] 0.54 [0.50-0.57] 0.69 [0.62-0.76] 0.65 [0.61-0.70] 0.67 [0.63-0.71] 0.61 [0.53-0.69] 0.50 [0.38-0.62]
PCA-OC SOCMT 0.40 [0.35-0.45] 0.79 [0.77-0.81] 0.36 [0.31-0.42] 0.77 [0.69-0.84] 0.29 [0.25-0.34] 0.76 [0.69-0.82] 0.29 [0.23-0.35] 0.79 [0.72-0.86]
Parzen-OC SOCMT 0.44 [0.39-0.50] 0.70 [0.64-0.76] 0.49 [0.46-0.52] 0.73 [0.67-0.79] 0.51 [0.47-0.54] 0.77 [0.70-0.83] 0.49 [0.45-0.52] 0.79 [0.70-0.87]
SVM-OC SOCMT 0.20 [0.16-0.24] 0.72 [0.68-0.77] 0.22 [0.20-0.25] 0.76 [0.70-0.81] 0.27 [0.24-0.30] 0.77 [0.70-0.83] 0.28 [0.25-0.31] 0.80 [0.72-0.88]
s K-NN-OC MOCV 0.99 [0.98-1.00] 0.01 [0.00-0.02] 1.00 [1.00-1.00] 0.00 [0.00-0.00] 1.00 [1.00-1.00] 0.00 [0.00-0.00] 1.00 [1.00-1.00] 0.00 [0.00-0.00]
KMeans-OC MOCV 0.36 [0.29-0.43] 0.71 [0.63-0.80] 0.48 [0.43-0.54] 0.78 [0.72-0.83] 0.59 [0.57-0.61] 0.69 [0.64-0.73] 0.46 [0.44-0.49] 0.68 [0.60-0.77]
PCA-OC MOCV 0.42 [0.36-0.48] 0.71 [0.65-0.77] 0.50 [0.48-0.52] 0.78 [0.72-0.83] 0.49 [0.46-0.53] 0.77 [0.70-0.83] 0.43 [0.40-0.45] 0.70 [0.60-0.80]
Parzen-OC MOCV 0.36 [0.31-0.40] 0.70 [0.64-0.76] 0.34 [0.28-0.41] 0.78 [0.72-0.83] 0.32 [0.28-0.36] 0.77 [0.70-0.83] 0.22 [0.20-0.24] 0.70 [0.60-0.80]
SVM-OC MOCV 0.19 [0.15-0.23] 0.76 [0.74-0.79] 0.25 [0.22-0.28] 0.79 [0.73-0.85] 0.27 [0.24-0.30] 0.78 [0.72-0.84] 0.22 [0.20-0.24] 0.72 [0.63-0.81]
  1. For each group of experiments results are provided in terms of average sensitivity (Sens) and specificity (Spec) with the related Confidence Intervals (CI). Confidence intervals are computed with 95% LOC.