Table 2

The effect of different machine learning algorithms on model prediction performance (bootstrapping)

	AUC		Accuracy		Precision		Recall rate		F1 value
	Mean±SD	95% CI	Mean±SD	95% CI	Mean±SD	95% CI	Mean±SD	95% CI	Mean±SD	95% CI
Machine learning algorithms
AdaBoost	0.702±0.104	0.700 to 0.703	0.761±0.061	0.760 to 0.762	0.434±0.134	0.432 to 0.436	0.538±0.142	0.535 to 0.540	0.465±0.105	0.463 to 0.467
Bagging	0.749±0.083	0.748 to 0.750	0.776±0.064	0.774 to 0.777	0.457±0.137	0.454 to 0.459	0.486±0.159	0.483 to 0.489	0.452±0.112	0.450 to 0.454
Bernoulli NB	0.718±0.099	0.716 to 0.720	0.771±0.056	0.770 to 0.772	0.444±0.133	0.442 to 0.447	0.541±0.141	0.538 to 0.543	0.475±0.109	0.474 to 0.477
DT	0.667±0.085	0.665 to 0.668	0.738±0.067	0.737 to 0.739	0.388±0.127	0.386 to 0.390	0.491±0.151	0.489 to 0.494	0.417±0.105	0.416 to 0.419
Ensemble Learning	0.793±0.083	0.791 to 0.794	0.810±0.058	0.809 to 0.811	0.545±0.157	0.543 to 0.548	0.576±0.162	0.573 to 0.579	0.537±0.108	0.535 to 0.539
ET	0.596±0.097	0.594 to 0.598	0.703±0.081	0.701 to 0.704	0.308±0.149	0.305 to 0.310	0.393±0.186	0.390 to 0.396	0.326±0.139	0.324 to 0.329
Gaussian NB	0.667±0.106	0.665 to 0.669	0.720±0.061	0.719 to 0.721	0.364±0.106	0.362 to 0.366	0.543±0.133	0.541 to 0.545	0.429±0.103	0.427 to 0.431
Gradient boosting	0.718±0.100	0.716 to 0.720	0.783±0.060	0.782 to 0.784	0.487±0.161	0.484 to 0.490	0.524±0.144	0.521 to 0.526	0.481±0.105	0.479 to 0.483
KNN	0.655±0.101	0.654 to 0.657	0.741±0.086	0.740 to 0.743	0.394±0.262	0.389 to 0.399	0.355±0.217	0.351 to 0.359	0.316±0.166	0.313 to 0.319
LDA	0.724±0.097	0.722 to 0.725	0.770±0.065	0.769 to 0.772	0.457±0.149	0.454 to 0.459	0.561±0.141	0.558 to 0.564	0.487±0.110	0.485 to 0.489
LR	0.728±0.094	0.727 to 0.730	0.770±0.070	0.769 to 0.771	0.465±0.155	0.462 to 0.467	0.580±0.143	0.577 to 0.583	0.497±0.110	0.495 to 0.499
Multinomial NB	0.727±0.099	0.725 to 0.728	0.753±0.071	0.752 to 0.754	0.450±0.170	0.447 to 0.453	0.570±0.175	0.567 to 0.573	0.467±0.111	0.465 to 0.469
Passive aggressive	0.686±0.094	0.684 to 0.688	0.701±0.087	0.699 to 0.703	0.358±0.119	0.355 to 0.360	0.558±0.156	0.555 to 0.560	0.421±0.107	0.419 to 0.423
QDA	0.660±0.115	0.658 to 0.662	0.774±0.057	0.773 to 0.775	0.428±0.178	0.425 to 0.431	0.436±0.188	0.433 to 0.440	0.411±0.152	0.408 to 0.413
RF	0.742±0.088	0.741 to 0.744	0.792±0.075	0.791 to 0.793	0.534±0.194	0.531 to 0.538	0.430±0.155	0.427 to 0.432	0.444±0.119	0.441 to 0.446
SGD	0.720±0.099	0.718 to 0.722	0.762±0.064	0.761 to 0.764	0.452±0.196	0.448 to 0.455	0.507±0.213	0.503 to 0.511	0.434±0.141	0.432 to 0.437
SVM	0.735±0.090	0.734 to 0.737	0.792±0.073	0.790 to 0.793	0.533±0.194	0.529 to 0.536	0.443±0.165	0.440 to 0.446	0.449±0.115	0.447 to 0.451
XGBoost	0.740±0.095	0.738 to 0.741	0.790±0.074	0.789 to 0.792	0.515±0.161	0.512 to 0.518	0.513±0.165	0.510 to 0.516	0.486±0.112	0.484 to 0.488
p value	p<0.0001		p<0.0001		p<0.0001		p<0.0001		p<0.0001

AUC, area under curve; DT, Decision Tree; ET, Extra Tree; KNN, K-Nearest Neighbour; LDA, Latent Dirichlet Allocation; LR, Logistic Regression; NB, Naïve Bayes; QDA, Quadratic Discriminant Analysis; SGD, Stochastic Gradient Descent; SVM, Support Vector Machine.