Skip to main content

Table 3 Observed differences between the testing results and each race with p values from permutation tests

From: Comparison between machine learning methods for mortality prediction for sepsis patients with different social determinants

 

Asian

Black or African American

Hispanic or Latino

Other

White

 

Observed difference

p_val

Observed difference

p_val

Observed difference

p_val

Observed difference

p_val

Observed difference

p_val

Ridge classifier

− 0.2812

0.009

− 0.0241

0.366

− 0.2208

0.038

0.0011

0.528

0.0175

0.286

Perceptron

− 0.2748

0.007

0.0078

0.448

− 0.2453

0.025

− 0.0026

0.502

0.0158

0.312

Passive-aggressive

− 0.3188

0.003

− 0.0075

0.464

− 0.1749

0.083

0.0159

0.381

0.0111

0.370

kNN

− 0.1314

0.141

− 0.0628

0.219

− 0.1865

0.069

− 0.0144

0.372

0.0214

0.245

Random forest

− 0.0834

0.207

− 0.0939

0.046

− 0.1226

0.112

0.0536

0.120

0.0056

0.429

LinearSVC_L1

− 0.2819

0.012

− 0.0172

0.410

− 0.2247

0.039

0.0003

0.481

0.0173

0.285

LinearSVC_L2

− 0.2815

0.009

− 0.0221

0.385

− 0.2211

0.045

0.0005

0.490

0.0175

0.294

SGDClassifier_L1

− 0.2872

0.008

− 0.0041

0.482

− 0.2159

0.044

0.0036

0.478

0.0184

0.266

SGDClassifier_L2

− 0.2900

0.008

− 0.0087

0.455

− 0.2182

0.039

0.0044

0.469

0.0191

0.263

SGDClassifier_EN

− 0.2905

0.010

− 0.0046

0.461

− 0.2186

0.058

0.0050

0.497

0.0181

0.300

MultinomialNB

− 0.2797

0.010

0.0671

0.182

− 0.2373

0.033

0.0051

0.484

0.0051

0.416

BernoulliNB

− 0.1974

0.025

− 0.0034

0.483

0.0476

0.331

0.0173

0.368

0.0012

0.490

Logistic regression

− 0.2875

0.010

− 0.0257

0.377

− 0.2061

0.054

0.0043

0.495

0.0174

0.273

SVC_rbf

− 0.3085

0.005

0.0042

0.480

− 0.2311

0.031

− 0.0176

0.383

0.0175

0.275

SVC_poly

− 0.2978

0.006

0.0027

0.483

− 0.2751

0.017

0.0087

0.431

0.0154

0.287

SVC_sigmoid

− 0.1343

0.144

− 0.0941

0.083

− 0.0606

0.332

− 0.0099

0.455

0.0208

0.247

  1. Observe difference: observed difference in AUC when compared with the performance on the entire testing set; p_val: p value, p values less than or equal to 0.05 were highlighted; Passive-aggressive: passive-aggressive classifier; kNN: k-Nearest Neighbors; LinearSVC_L1 or _L2: support vector machine with linear kernel coupled with L1 or L2 regularization; SGDClassifier_L1 or _L2 or _EN: stochastic gradient descent with L1 or L2 or Elastic Net regularization; MultinomialNB: Multinomial naïve Bayes; BernoulliNB: Bernoulli naïve Bayes; SVC_rbf or _poly or _sigmoid: support vector machine with rbf kernel or polynomial kernel or sigmoid kernel