Predicting Flight Delays with Machine Learning: A Case Study from Saudi Arabian Airlines

Figure 3

The results of the confusion matrices, which provide a visual representation of the performance of each model in terms of correctly and incorrectly classified instances across our different classes (on time and delayed).

(a) CatBoost confusion matrix

(b) XGBoost confusion matrix

(c) LightGBM confusion matrix

(d) MLP confusion matrix

(e) Random forest confusion matrix