| | Studies | Data | Algorithm | Evaluation method | Performance | |
| | Denney et al. [23] | 7 million | Ada, LR, SVM, NB, SGD, ET, DT, XG, RF | Average recall | 68% recall | Existing model with results | | AlMuhaideb et al. [29] | 1.1 million | JRip, Hoeffding trees, LR, MP, NB | Accuracy, AUC | 77.13% accuracy, 0.86 AUC | | Mohammadi et al. [30] | 74 thousand | LR, MP, NB | Accuracy, AUC | 82% accuracy, 0.86 AUC | | Daghistani et al. [31] | 201 million | RF, GB, LR, SVM, MP | Accuracy, precision, recall, F1 measure, AUC | 79% accuracy, 77% precision, 79% recall, 76% score, 0.81 AUC | | Our model | 6 million | Ada, LR, SVM, NB, SGD, XG, DT, GB, RF, MP | Accuracy, precision, recall, F1 measure, AUC, MSE | 86.5% accuracy, 83% precision, 94% recall, 87% score, 0.92 AUC, 0.1069 MSE | Proposed model with result |
|
|
