Skip to main content

Table 21 Maximum AUPRC by classifier for the data exfiltration attack type; mean and standard deviations of AUPRC, (10 iterations of 5-fold cross-validation)

From: IoT information theft prediction using ensemble feature selection

Experiment name Mean SD
AUPRC AUPRC
LR Tuned RUS 1:9 6 Agree 0.93654 0.06464
CatBoost 5 Agree 0.99448 0.00784
DT Tuned 6 Agree 0.96776 0.02877
Light GBM RUS 1:9 6 Agree 0.99079 0.01350
RF 4 Agree 0.99199 0.01341
MLP 4 Agree 0.95897 0.04293
NB Tuned RUS 1:9 7 Agree 0.84859 0.04161
XGBoost 5 Agree 0.98823 0.01938