Table 1.
The performance of different machine learning techniques based models on Antifp_Main dataset developed using amino acid composition of peptides.
| Parameter | Main Dataset |
Validation Dataset |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Sen | Spc | Acc | MCC | ROC | Sen | Spc | Acc | MCC | ROC | ||
| SVM | g = 0.01, c = 5, j = 4 | 88.61 | 87.93 | 88.27 | 0.77 | 0.94 | 86.60 | 85.91 | 86.25 | 0.73 | 0.94 |
| Random Forest | Ntree = 130 | 87.84 | 86.64 | 87.24 | 0.74 | 0.93 | 86.94 | 80.76 | 83.85 | 0.68 | 0.91 |
| SMO | g = 0.001, c = 2 | 87.84 | 82.11 | 84.97 | 0.70 | 0.84 | 88.32 | 81.44 | 84.88 | 0.70 | 0.84 |
| J48 | c = 0.1, m = 7 | 80.39 | 80.65 | 80.52 | 0.61 | 0.82 | 82.82 | 81.44 | 82.13 | 0.64 | 0.84 |
| Naïve Bayes | Default | 76.46 | 75.86 | 76.16 | 0.52 | 0.80 | 74.91 | 78.01 | 76.46 | 0.53 | 0.81 |
Sen, Sensitivity; Spc, Specificity; Acc, Accuracy; MCC, Matthews Correlation Coefficient; ROC, Receiver Operating Characteristic.