Table 8 Performance metrics: proposed model versus existing ones without balancing
Model, year with reference | Accuracy metrics (%) | |||||||
|---|---|---|---|---|---|---|---|---|
ACC | PR | RE | FS | |||||
Binary | 4 class | Binary | 4 class | Binary | 4 class | Binary | 4 class | |
MalHyStack, 2023 (Roy et al. 2023) | 99.85 | 85.04 | 99.97 | 85.04 | 99.73 | 85.17 | 99.85 | 84.96 |
RobustCBL, 2023 (Shafin et al. 2023) | 99.96 | 84.56 | 100.00 | 85.00 | 100.00 | 85.00 | 100.00 | 85.00 |
CatBoost, 2022 (Dang 2022) | 99.97 | 84.86 | 99.98 | 79.69 | 99.98 | 88.46 | 99.97 | 71.49 |
DT, 2022 (Mezina and Burget 2022) | 99.00 | 79.16 | 99.00 | 69.00 | 100.00 | 69.00 | 99.00 | 69.00 |
DCNN, 2022 (Mezina and Burget 2022) | 99.92 | 83.53 | 99.00 | 76.00 | 99.00 | 75.00 | 99.00 | 75.00 |
Proposed | 100.00 | 99.96 | 100.00 | 99.96 | 100.00 | 99.96 | 100.00 | 99.96 |