Skip to main content

Table 34 Classification results for the keylogging dataset with sampling and tuned hyperparameters experiments; mean and standard deviations of AUC and AUPRC, (10 iterations of 5-fold cross-validation)

From: IoT information theft prediction using ensemble feature selection

Experiment name Mean SD Mean SD
AUC AUC AUPRC AUPRC
CatBoost Tuned RUS 1:3 0.99709 0.00191 0.99968 0.00054
CatBoost Tuned RUS 1:1 0.99749 0.00170 0.99971 0.00051
DT Tuned RUS 1:3 0.99468 0.00347 0.99405 0.00362
DT Tuned RUS 1:1 0.99412 0.00270 0.99259 0.00661
Light GBM Tuned RUS 1:3 0.99675 0.00200 0.99932 0.00097
Light GBM Tuned RUS 1:1 0.99674 0.00198 0.99957 0.00066
LR Tuned RUS 1:3 0.99589 0.00240 0.99786 0.00223
LR Tuned RUS 1:1 0.99609 0.00214 0.99688 0.00992
MLP Tuned RUS 1:3 0.99489 0.00220 0.99785 0.00203
MLP Tuned RUS 1:1 0.99187 0.01955 0.99679 0.00440
NB Tuned RUS 1:3 0.99240 0.00182 0.97520 0.00536
NB Tuned RUS 1:1 0.99259 0.00208 0.97334 0.00680
RF Tuned RUS 1:3 0.99702 0.00194 0.99945 0.00077
RF Tuned RUS 1:1 0.99668 0.00210 0.99963 0.00047
XGBoost Tuned RUS 1:3 0.99640 0.00190 0.99956 0.00058
XGBoost Tuned RUS 1:1 0.99677 0.00184 0.99957 0.00057