Table 3.
Performance evaluation of various models under within-dataset environment. (a) Waseem & Hovy, (b) SemEval 2019. The first row for each dataset represents the Micro average and the second represents the Macro average.
| Method | Dataset | F1 | Prec. | Rec. |
|---|---|---|---|---|
| Proposed model | (a) | 80.59 | 80.56 | 80.87 |
| 74.20 |
75.61 |
73.15 |
||
| (b) | 73.97 | 74.06 | 73.95 | |
| 73.31 | 73.29 | 73.39 | ||
| Badjatiya et al. (2017) | (a) | 80.70 | 82.30 | 82.10 |
| 73.10 |
81.60 |
68.90 |
||
| (b) | 75.26 | 75.30 | 75.32 | |
| 74.54 | 74.73 | 74.46 | ||
| Agrawal and Awekar (2018) | (a) | 84.30 | 84.70 | 84.10 |
| 79.60 |
78.00 |
81.70 |
||
| (b) | 70.85 | 76.51 | 71.19 | |
| 71.81 | 74.40 | 70.96 | ||