Fig 1.
Flowchart of the proposed method.
Table 1.
Different features of the dataset.
Fig 2.
Schematic representation of the autoencoder.
Fig 3.
Schematic representation of a random forest.
Fig 4.
Schematic representation of XGBoost.
Table 2.
Confusion matrix.
Fig 5.
Schematic representation of a reconstructed error.
Table 3.
Comparison of different train-test splits.
Fig 6.
Experiment flow structure.
Fig 7.
XGBoost-AE training set results.
Fig 8.
XGBoost-AE test set results.
Fig 9.
XGBoost-AE unseen dataset results.
Fig 10.
Random Forest-AE training set results.
Fig 11.
Random Forest-AE test set results.
Fig 12.
Random Forest-AE unseen dataset results.
Fig 13.
Model evaluation on a 20% test set.
Table 4.
Model evaluation on a 20% test set.
Fig 14.
Previous research model evaluation on a 20% test set.
Table 5.
Results compared with previous research.
Fig 15.
Model evaluation on an unseen data test set.
Table 6.
Model evaluation on an unseen test set.