Research Article
Detection of COVID-19 Using Protein Sequence Data via Machine Learning Classification Approach
Table 6
Average specificity of different models for different training percentage.
| | Training percentage | GA+KNN | SVM-RFE+KNN | LASSO+KNN | GA+Naïve Bayes | SVM-RFE+Naïve Bayes | LASSO+Naïve Bayes | GA+XGBoost | SVM-RFE+XGBoost | LASSO+XGBoost |
| | 60% | 98.79% | 98.93% | 96.90% | 90.65% | 90.41% | 87.72% | 92.12% | 98.01% | 94.16% | | 70% | 98.38% | 98.40% | 96.61% | 91.38% | 90.65% | 87.98% | 95.24% | 98.01% | 94.44% | | 80% | 99.01% | 99.34% | 97.01% | 91.51% | 91.40% | 88.24% | 93.11% | 98.08% | 95.05% | | 90% | 99.04% | 99.52% | 96.87% | 91.90% | 91.76% | 88.16% | 93.44% | 96.08% | 93.84% |
|
|
Highest specificity achieved at a training percentage of 90% using the SVM-RFE+KNN model, reaching 99.52%, as observed from the boldfaced entries.
|
