Hands-On Artificial Intelligence for Cybersecurity

The XGBClassifier report

The classification report for the XGBClassifier metrics is shown in the following code:

print('classification report')print(metrics.classification_report(ytest, y_pred))print("Accuracy is :")print(metrics.accuracy_score(ytest, y_pred))print('Area under the curve : %f' % (metrics.roc_auc_score(ytest, y_pred)))classification report             precision    recall  f1-score   support          0       1.00      1.00      1.00     17030          1       0.93      0.79      0.85        33avg / total       1.00      1.00      1.00     17063Accuracy is : 0.999472542929Area under the curve : 0.893881

By comparing the AUC and F1 score values, which are calculated using the individual classifiers, XGBClassifier remains the most accurate classifier and GradientBoostingClassifier is the least accurate of the three.