Table 8 The results of the proposed model with different ML algorithms using LSTM with multi-attention features.
From: A hybrid super learner ensemble for phishing detection on mobile devices
Metrics | ML1 | ML2 | ML3 | ML4 | ML5 |
|---|---|---|---|---|---|
Recall (%) | 98.29 | 98.52 | 98.18 | 98.43 | 98.37 |
TNR (%) | 97.99 | 98.64 | 98.37 | 98.13 | 98.37 |
Precision (%) | 98.51 | 98.99 | 98.80 | 98.62 | 98.79 |
F1-Score | 98.40 | 98.76 | 98.49 | 98.53 | 98.58 |
Accuracy (%) | 98.16 | 98.57 | 98.26 | 98.31 | 98.37 |
MCC (%) | 96.25 | 97.08 | 96.45 | 96.54 | 96.67 |
