Skip to main content
. 2023 Jun 27;14:1170459. doi: 10.3389/fendo.2023.1170459

Table 2.

Classification 2 (C-2) performance (Patients with NGT vs. Patients with T2D).

ML algorithms Accuracy AUC Mean Accuracy
(SD, CV=10)
Mean AUC
(SD, CV=10)
Logistic Regression 1 1 0.94 (0.04) 0.94 (0.06)
Naive Bayes 0.87 0.84 0.85 (0.04) 0.92 (0.04)
Decision Tree 0.86 0.83 0.84 (0.08) 0.75 (0.16)
Random Forest 0.89 0.98 0.96 (0.03) 0.99 (0.02)
XGBoost 0.92 0.96 0.91 (0.04) 0.98 (0.02)
Multilayer Perceptron (MLP) 1 1 0.91 (0.07) 0.93 (0.07)

We compared the performance of six ML algorithms using the Precision and area under the receiver operating characteristic (ROC) curve (AUC) values. We use the stratified cross-validation (CV) technique to obtain our results’ standard deviation (SD) (K Fold = 10).