Skip to main content

Table 3 Performance comparison of classifiers under different percentage of pseudo-negative samples on the Haberman data

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

PercentageClassifierSen(%)Spe(%)Acc(%)MCC
0DA17.3395.4274.790.212
 AdaBoost29.1990.8974.710.266
 RF34.282.8470.070.197
 NN27.9887.2871.680.202
10DA21.7793.9672.960.236
 AdaBoost32.7286.1270.580.214
 RF33.3883.9169.380.197
 NN30.3782.0167.040.144
20DA30.5194.4174.20.340
 AdaBoost46.6887.5474.260.370
 RF45.0181.3269.590.272
 NN37.4282.9768.570.222
30DA31.7395.173.320.36
 AdaBoost51.8187.1575.650.422
 RF51.0679.6700.311
 NN42.3984.5470.360.291
40DA37.1394.3872.930.404
 AdaBoost50.7386.172.870.396
 RF56.8178.3869.950.359
 NN53.638170.60.35
50DA38.6193.8371.740.405
 AdaBoost61.4682.2673.810.447
 RF60.7578.2270.950.395
 NN52.4179.8168.560.339