Table 2: Overall performance results of the proposed classification approach.

Sensitivity (%)TCA (%)Specificity (%)AUROCKappaRMSE

Method 1
 ANN87.584.983.80.9280.6930.356
 RF87.580.281.80.8890.5940.375
 Bagging (ANN)10061.61000.5990.1450.557
 Bagging (RF)87.586.084.20.9110.7170.339
 Adaboost (ANN)87.581.482.40.9200.6190.416
 Adaboost (RF)83.375.675.80.8800.4980.384

Method 2
 ANN85.483.781.60.9130.6700.327
 RF83.383.780.00.8900.6720.356
 Bagging (ANN)10060.51000.6160.1160.522
 Bagging (RF)83.381.478.90.9120.6230.346
 Adaboost (ANN)81.380.276.90.8690.6000.385
 Adaboost (RF)83.380.278.40.9010.5980.351

Method 3
 ANN83.383.780.00.8930.6720.358
 RF81.377.975.70.8690.5510.383
 Bagging (ANN)39.659.352.50.7380.2240.482
 Bagging (RF)83.384.980.50.9120.6960.344
 Adaboost (ANN)83.383.780.00.8670.6720.378
 Adaboost (RF)85.483.781.60.9060.6700.348

Method 4
 ANN89.690.787.50.9400.8120.307
 RF85.483.781.60.9080.6700.335
 Bagging (ANN)60.466.359.60.7700.3330.463
 Bagging (RF)87.586.084.20.9220.7170.325
 Adaboost (ANN)87.586.084.20.9210.7170.334
 Adaboost (RF)85.480.280.00.9110.5960.343