Research Article
Detection of COVID-19 Using Protein Sequence Data via Machine Learning Classification Approach
Table 5
Average accuracy of different models for different training percentages.
| | 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% | 96.50% | 98.55% | 97.29% | 81.21% | 80.85% | 79.37% | 93.91% | 97.25% | 93.66% | | 70% | 96.30% | 98.99% | 96.92% | 81.55% | 81.32% | 80.35% | 93.15% | 97.33% | 93.04% | | 80% | 96.50% | 99.19% | 96.99% | 82.42% | 81.45% | 81.44% | 93.33% | 97.50% | 92.12% | | 90% | 97.50% | 99.30% | 96.54% | 82.19% | 81.07% | 80.03% | 92.87% | 96.00% | 92.10% |
|
|
Highest accuracy achieved at a training percentage of 90% using the SVM-RFE+KNN model, reaching 99.30%, as observed from the boldfaced entries.
|
