Table 2. Performance of five machine learning algorithms with L1O cross validation using 46 CRKP and 49 CSKP mass spectra, in terms of accuracy, sensitivity and specificity for differentiate CRKP from CSKP.
| Algorithm | Metric | No. of ranked peaks selected with increasing p-value | |||||
|---|---|---|---|---|---|---|---|
| 50 | 60 | 70 | 80 | 90 | 100 | ||
| Random Forest |
Accuracy | 94% | 94% | 95% | 97% | 95% | 94% |
| Sensitivity | 91% | 91% | 91% | 93% | 89% | 87% | |
| Specificity | 96% | 96% | 98% | 100% | 100% | 100% | |
| Logistic Regression |
Accuracy | 93% | 92% | 92% | 93% | 91% | 91% |
| Sensitivity | 93% | 93% | 96% | 98% | 96% | 98% | |
| Specificity | 92% | 90% | 88% | 88% | 86% | 84% | |
| Naïve Bayes |
Accuracy | 86% | 88% | 87% | 89% | 88% | 89% |
| Sensitivity | 74% | 76% | 74% | 78% | 76% | 78% | |
| Specificity | 98% | 100% | 100% | 100% | 100% | 100% | |
| Nearest Neighbors |
Accuracy | 91% | 84% | 87% | 87% | 83% | 85% |
| Sensitivity | 87% | 83% | 87% | 89% | 87% | 91% | |
| Specificity | 94% | 86% | 88% | 86% | 80% | 80% | |
| Support Vector Machine |
Accuracy | 87% | 84% | 84% | 86% | 85% | 87% |
| Sensitivity | 96% | 96% | 96% | 100% | 98% | 98% | |
| Specificity | 80% | 73% | 73% | 73% | 73% | 78% | |