Table 1.
Five-fold cross validation statistics for all SARS-CoV-2 machine learning models implemented using ECFP6 fingerprints.
| ACC | AUC | CK | MCC | Pr | Recall | Sp | F1 | |
|---|---|---|---|---|---|---|---|---|
| AC | 0.81 | 0.78 | 0.62 | 0.64 | 0.78 | 0.88 | 0.73 | 0.83 |
| rf | 0.75 | 0.74 | 0.49 | 0.5 | 0.73 | 0.82 | 0.67 | 0.77 |
| knn | 0.71 | 0.71 | 0.43 | 0.42 | 0.71 | 0.76 | 0.67 | 0.74 |
| svc | 0.7 | 0.69 | 0.39 | 0.4 | 0.68 | 0.79 | 0.6 | 0.73 |
| bnb | 0.68 | 0.68 | 0.36 | 0.36 | 0.7 | 0.7 | 0.67 | 0.7 |
| ada | 0.64 | 0.63 | 0.27 | 0.26 | 0.65 | 0.67 | 0.6 | 0.66 |
| DL | 0.65 | 0.65 | 0.3 | 0.3 | 0.66 | 0.67 | 0.63 | 0.66 |
ACC: Accuracy, AUC: Area under curve, CK: Cohen’s Kappa, MCC: Matthews correlation coefficient, Pr: Precision, Sp: Specificity, F1: F1 Score. bnb: Bernoulli Naïve Bayes, ada: AdaBoost Decision trees, rf: Random Forest, svc: support vector machine classifier, knn: k-Nearest Neighbors and DL: Deep Learning (DL).