Table 11.
Performance comparison between the proposed methods.
| Method Class | Method | Precision | Recall | F1-Score | 
|---|---|---|---|---|
| Linear methods | UTurku [16] | 0.732 | 0.499 | 0.594 | 
| UWM [36] | 0.439 | 0.505 | 0.47 | |
| Kim [37] | - | - | 0.67 | |
| Kernel methods | Nil [38] | 0.535 | 0.501 | 0.517 | 
| WBI [39] | 0.642 | 0.579 | 0.609 | |
| FBK-irst [17] | - | - | 0.67 | |
| RNN Methods | Liu et al. [6] | 0.757 | 0.647 | 0.698 | 
| Quan et al. [18] | 0.76 | 0.653 | 0.702 | |
| Park et al. [20] | 0.6895 | 0.7477 | 0.7174 | |
| Zhang et al. [34] | 0.741 | 0.718 | 0.729 | |
| Wuti et al. [40] | 0.801 | 0.740 | 0.770 | |
| This work | Proposed Model No. 1 | 0.7359 | 0.7405 | 0.7382 | 
| Proposed Model No. 2 | 0.837 | 0.85 | 0.8438 |