Fig. 5

Comparison of performance of different machine learning models to predict BTCP events using 5-folds cross-validation on 80% of the dataset (475 out of the total 594 observations) is shown. Each deployment was trained and validated on their own data. Note: NB = Naïve Bayes; DT = Decision Tree; RF = Random Forest; SVM = Support Vector Machine; NN = Neural Network; AUC = Area Under Curve; MCC = Matthew’s correlation coefficient