Skip to main content

Table 19 Classification results for the data exfiltration 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 1:9

0.99281

0.00986

0.99292

0.01192

CatBoost Tuned RUS 1:3

0.98816

0.01388

0.98791

0.01652

CatBoost Tuned RUS 1:1

0.98728

0.01193

0.98444

0.01864

DT Tuned RUS 1:9

0.96722

0.03625

0.91543

0.07238

DT Tuned RUS 1:3

0.96993

0.03208

0.84267

0.08491

DT Tuned RUS 1:1

0.97706

0.01870

0.74923

0.08163

Light GBM Tuned RUS 1:9

0.98769

0.01770

0.98951

0.01959

Light GBM Tuned RUS 1:3

0.98960

0.01336

0.95978

0.08703

Light GBM Tuned RUS 1:1

0.98446

0.01508

0.91785

0.09834

LR Tuned RUS 1:9

0.95866

0.03334

0.90807

0.10925

LR Tuned RUS 1:3

0.96372

0.03201

0.81231

0.15087

LR Tuned RUS 1:1

0.96302

0.02723

0.67686

0.15659

MLP Tuned RUS 1:9

0.87040

0.11157

0.93840

0.07158

MLP Tuned RUS 1:3

0.95885

0.03652

0.85402

0.14352

MLP Tuned RUS 1:1

0.96377

0.02815

0.72101

0.15642

NB Tuned RUS 1:9

0.97481

0.02199

0.81701

0.05655

NB Tuned RUS 1:3

0.97188

0.02087

0.70621

0.06589

NB Tuned RUS 1:1

0.96574

0.02068

0.64355

0.06437

RF Tuned RUS 1:9

0.98592

0.01871

0.98808

0.01699

RF Tuned RUS 1:3

0.98663

0.01519

0.98461

0.01831

RF Tuned RUS 1:1

0.98708

0.01384

0.97930

0.02606

XGBoost Tuned RUS 1:9

0.98292

0.02143

0.98610

0.02217

XGBoost Tuned RUS 1:3

0.98713

0.01487

0.97596

0.03204

XGBoost Tuned RUS 1:1

0.98634

0.01371

0.93328

0.08744