Table 2 Kappa metric results after 30-fold cross-validation. It compares the accuracy of the models to the accuracy of a random system. This statistic is a measure that can handle both multiclass and imbalanced class problems.
From: Machine learning random forest for predicting oncosomatic variant NGS analysis
Kappa results | |||||||
|---|---|---|---|---|---|---|---|
Min. | 1st Qu. | Median | Mean | 3rd Qu. | Max | NA’s | |
lda | 0.6863514 | 0.6936705 | 0.7043258 | 0.7016283 | 0.7090152 | 0.7125000 | 0 |
cart | 0.8338433 | 0.8453000 | 0.8816417 | 0.8813644 | 0.9206570 | 0.9277045 | 0 |
knn | 0.9156957 | 0.9162500 | 0.9214557 | 0.9213574 | 0.9261705 | 0.9274688 | 0 |
svm | 0.9387329 | 0.9447439 | 0.9462042 | 0.9455949 | 0.9469992 | 0.9490259 | 0 |
rf | 0.9948536 | 0.9949153 | 0.9955885 | 0.9957113 | 0.9960788 | 0.9973043 | 0 |
