Skip to main content
Journal of Big Data

Table 5 Case 2: SlowlorisBig results

From: Severely imbalanced Big Data challenges: investigating data sampling approaches

Learner

Method

(All:all)

(99:1)

(90:10)

(75:25)

(65:35)

(50:50)

(a) AUC

 GBT

None

0.68678

–

–

–

–

–

RUS

–

0.84644

0.97226

0.96541

0.96724

0.96685

ROS

–

0.65312

0.50947

0.69950

0.66151

0.65531

ADASYN

–

0.47154

0.74951

0.68449

0.82351

0.46056

SMOTE

–

0.57069

0.58314

0.69230

0.63663

0.70906

SMOTEb1

–

0.70283

0.65169

0.62276

0.62191

0.67668

SMOTEb2

–

0.69876

0.62302

0.63359

0.66083

0.71559

 LR

None

0.59203

–

–

–

–

–

RUS

–

0.62018

0.84740

0.90919

0.97113

0.95052

ROS

–

0.59869

0.63752

0.60610

0.60996

0.62989

ADASYN

–

0.77948

0.49306

0.43431

0.43311

0.46447

SMOTE

–

0.60657

0.64287

0.61986

0.62587

0.61301

SMOTEb1

–

0.59232

0.59301

0.59212

0.59242

0.59190

SMOTEb2

–

0.59257

0.59164

0.59254

0.59189

0.59143

 RF

None

0.86773

–

–

–

–

–

RUS

–

0.88343

0.88444

0.91207

0.95425

0.96045

ROS

–

0.88391

0.90186

0.91679

0.93715

0.95694

ADASYN

–

0.75805

0.68151

0.48584

0.46859

0.40685

SMOTE

–

0.88436

0.89994

0.91701

0.94070

0.95690

SMOTEb1

–

0.87027

0.85896

0.88157

0.87659

0.86275

SMOTEb2

–

0.87138

0.88829

0.86941

0.87098

0.86720

(b) GM

 GBT

None

0.25168

–

–

–

–

–

RUS

–

0.67700

0.83073

0.90015

0.94949

0.96174

ROS

–

0.48453

0.34405

0.53269

0.52886

0.49330

ADASYN

–

0.24369

0.31263

0.16892

0.31140

0.10138

SMOTE

–

0.47393

0.34644

0.59461

0.44976

0.57714

SMOTEb1

–

0.29552

0.30041

0.30273

0.32697

0.26873

SMOTEb2

–

0.28809

0.26814

0.27291

0.27253

0.28508

 LR

None

0.64486

–

–

–

–

–

RUS

–

0.62135

0.82268

0.90304

0.96983

0.94733

ROS

–

0.66742

0.72338

0.69552

0.70111

0.72352

ADASYN

–

0.76272

0.41830

0.37992

0.38049

0.41591

SMOTE

–

0.64121

0.72341

0.71979

0.72363

0.68929

SMOTEb1

–

0.64057

0.64489

0.64421

0.64489

0.64038

SMOTEb2

–

0.63685

0.63808

0.64065

0.64097

0.63461

 RF

None

0

–

–

–

–

–

RUS

–

0

0.15255

0.56097

0.65159

0.90195

ROS

–

0

0.38138

0.42997

0.64770

0.65151

ADASYN

–

0

0

0

0

0

SMOTE

–

0

0.34325

0.53447

0.64772

0.65155

SMOTEb1

–

0

0

0

0

0

SMOTEb2

–

0

0

0

0

0

  1. The highest value within each column (class distribution ratio) of each sub-table is in italic type, and the highest value within each row (sampling method) of each sub-table is underlined
Back to article page