TABLE 2.
Accuracy, sensitivity, and specificity obtained by cross-validation for responses to standard, deviant stimuli; k is the number of neighbors used in the classification algorithm.
k = 3 | k = 4 | k = 6 | k = 8 | SVM | Gaussian | |
Accuracy | ||||||
Standard responses | 0.96 | 0.96 | 0.89 | 0.85 | 1.0 | 0.92 |
Deviant responses | 0.89 | 0.89 | 0.89 | 0.89 | 1.0 | 0.89 |
Sensitivity | ||||||
Standard responses | 0.83 | 0.83 | 0.83 | 0.66 | 1.0 | 0.66 |
Deviant responses | 0.83 | 0.83 | 0.83 | 0.83 | 1.0 | 0.5 |
Specificity | ||||||
Standard responses | 1.0 | 1.0 | 0.9 | 0.9 | 1.0 | 1.0 |
Deviant responses | 0.91 | 0.91 | 0.91 | 0.91 | 1.0 | 1.0 |