Skip to main content

Table 2 Performance comparison of classifiers under different percentage of pseudo-negative samples on the CMC data

From: How to balance the bioinformatics data: pseudo-negative sampling

PercentageClassifierSen(%)Spe(%)Acc(%)MCC
0DA9.3897.8177.80.156
 AdaBoost21.3794.4877.940.226
 RF28.1992.878.20.27
 NN27.0187.0973.520.161
10DA17.694.8575.70.198
 AdaBoost25.7693.5876.780.266
 RF39.2291.7778.750.369
 NN40.9286.9875.550.302
20DA37.3591.7176.940.351
 AdaBoost40.0391.2477.330.36
 RF43.9491.2478.410.404
 NN47.2887.2276.380.368
30DA52.4688.3477.80.438
 AdaBoost50.8988.8377.670.431
 RF50.8789.9878.480.448
 NN53.3987.8677.730.439
40DA59.4687.2178.430.485
 AdaBoost56.0187.6177.610.461
 RF56.4590.2779.570.505
 NN54.9486.6876.640.439
50DA66.7885.4279.080.530
 AdaBoost64.0187.3779.420.531
 RF6288.7179.630.532
 NN61.0287.3878.420.505