Table 2.
Performance comparison of the baseline models and our model (GSCCAM) by four doctors for human qualitative evaluation.
| Models | Doctor1 | Doctor2 | Doctor3 | Doctor4 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HE | HC | Total | HE | HC | Total | HE | HC | Total | HE | HC | Total | |
| KDHR [18] | 2.2 | 1.5 | 3.7 | 2.3 | 1.1 | 3.4 | 2.0 | 1.2 | 3.2 | 2.4 | 1.6 | 4.0 |
| PTM [21] | 2.9 | 1.7 | 4.6 | 2.8 | 1.9 | 4.7 | 2.6 | 1.5 | 4.1 | 3.1 | 2.0 | 5.1 |
| Coverage-Soft-Model [44] | 1.9 | 1.5 | 3.4 | 2.2 | 1.4 | 3.6 | 2.0 | 1.3 | 3.3 | 2.5 | 1.5 | 4.0 |
| Herb-Know [24] | 3.5 | 1.8 | 5.3 | 3.1 | 2.3 | 5.4 | 2.9 | 1.9 | 4.8 | 3.8 | 2.2 | 6.0 |
| TPGen [23] | 3.7 | 1.8 | 5.5 | 3.2 | 2.2 | 5.4 | 2.9 | 2.1 | 5.0 | 3.6 | 2.5 | 6.1 |
| GSCCAM | 4.3 | 3.5 | 7.8 | 4.2 | 3.4 | 7.6 | 3.9 | 3.6 | 7.5 | 4.2 | 3.7 | 7.9 |