From: Predicting drug−disease associations via sigmoid kernel-based convolutional neural networks
Test set | Acc. (%) | Pre. (%) | Recall (%) | F1-score (%) |
---|---|---|---|---|
1 | 86.08 | 83.33 | 90.21 | 86.63 |
2 | 83.51 | 83.51 | 83.51 | 83.51 |
3 | 84.54 | 81.31 | 89.69 | 85.29 |
4 | 81.35 | 82.70 | 79.27 | 80.95 |
5 | 82.64 | 82.14 | 83.42 | 82.78 |
6 | 84.20 | 84.38 | 83.94 | 84.16 |
7 | 82.90 | 83.96 | 81.35 | 82.63 |
8 | 82.38 | 78.28 | 89.64 | 83.57 |
9 | 83.42 | 81.16 | 87.05 | 84.00 |
10 | 86.53 | 85.79 | 87.56 | 86.67 |
Average | 83.76 ± 1.54 | 82.66 ± 1.98 | 85.56 ± 3.61 | 84.02 ± 1.70 |
SKCNN | 89.55 ± 1.15 | 91.65 ± 1.77 | 87.07 ± 1.75 | 89.28 ± 1.19 |