Table 3 Predictive results of 5-fold cross-validation performed by the two models on Human dataset
From: Using discriminative vector machine model with 2DPCA to predict interactions among proteins
Model | Test set | Acc (%) | Sen (%) | Pre (%) | MCC |
|---|---|---|---|---|---|
DVM | 1 | 97.86 | 98.06 | 96.57 | 0.9473 |
2 | 97.43 | 97.37 | 95.50 | 0.9393 | |
3 | 97.04 | 97.73 | 96.41 | 0.9401 | |
4 | 97.98 | 97.89 | 98.07 | 0.9495 | |
5 | 97.80 | 97.51 | 96.61 | 0.9462 | |
Average | 97.62 ± 0.38 | 97.71 ± 0.28 | 96.63 ± 0.92 | 0.9445 ± 0.0045 | |
SVM | 1 | 93.79 | 93.40 | 93.52 | 0.8855 |
2 | 92.69 | 94.06 | 91.15 | 0.8642 | |
3 | 93.42 | 91.44 | 92.57 | 0.8780 | |
4 | 92.93 | 90.78 | 94.33 | 0.8688 | |
5 | 93.18 | 93.30 | 92.95 | 0.8736 | |
Average | 93.20 ± 0.43 | 92.60 ± 1.41 | 92.90 ± 1.18 | 0.8740 ± 0.0082 |