Table 3 Forecast results for testing group.

From: A predictive model for post-thoracoscopic surgery pulmonary complications based on the PBNN algorithm

Model name

AUC

Accuracy

Precision

Recall

F1

Mse

MCC

Specificity

Logistic regression

0.850

0.831

0.660

0.532

0.589

0.169

0.489

0.919

Decision tree classifier

0.842

0.816

0.658

0.403

0.500

0.184

0.413

0.938

Random forest classifier

0.855

0.779

0.625

0.081

0.143

0.221

0.165

0.986

Gradient boosting classifier

0.839

0.809

0.750

0.242

0.366

0.191

0.351

0.976

XGB classifier

0.852

0.809

0.625

0.403

0.490

0.191

0.435

0.971

LGBM classifier

0.873

0.827

0.742

0.371

0.495

0.173

0.439

0.962

LinearSVC

0.845

0.816

0.667

0.387

0.490

0.184

0.408

0.943

MLPC

0.864

0.801

0.618

0.339

0.438

0.199

0.351

0.938

gnb

0.811

0.787

0.529

0.581

0.554

0.213

0.415

0.848

knn

0.747

0.776

0.520

0.21

0.299

0.224

0.221

0.943

adab

0.781

0.816

0.630

0.468

0.537

0.184

0.433

0.919

CNN

0.819

0.794

0.571

0.387

0.462

0.206

0.334

0.914

LSTM

0.833

0.809

0.647

0.355

0.458

0.191

0.418

0.933

CNNRNN

0.775

0.783

0.528

0.452

0.487

0.217

0.345

0.857

CNNLSTM

0.786

0.787

0.556

0.323

0.408

0.213

0.308

0.871

PBNN

0.869

0.820

0.627

0.516

0.566

0.180

0.452

0.914

  1. Abbreviate: Logistic Regression, Decision Tree, Random Forest, Gradient Boosting Decision Tree-Gradient Boosting, Extreme gradient boosting-XGB, light gradient boosting machine-LGBM, Linear Support Vector-Linear SVC, Multilayer Perceptron Classifier-MLPC, Gaussian naive Bayes-gnb, K-nearst neighbors-knn, AdaBoost-adab, Convolutional Neural Network-CNN, Long Short Term Memory-LSTM, Convolutional Neural Network + Recurrent Neural Networks-CNNRNN, Convolutional Neural Network + Long Short Term Memory-CNNLSTM and Pruning Bayesian neural network-PBNN; Matthews correlation coefficient-MCC.
Back to article page