Table 1 Classification accuracy with different feature sets versus varying classifiers from the combination of all affective sources (A+L+D)
From: Quality of experience evaluation of voice communication: an affect-based approach
Classifier Model | Base | f10 | f15 | PCA |
|---|---|---|---|---|
SVM | 57.3 | 61.1 | 62.5 | 65.7 |
SVM-5WC | 58.7 | 62.3 | 62.7 | 64.8 |
SVM-10WC | 58.9 | 61.5 | 62.4 | 65.2 |
SVM-5CV | 60.6 | 63.7 | 64.1 | 67.9 |
kNN | 49.7 | 52.8 | 53.1 | 54.3 |
kNN-5CV | 53.5 | 54.2 | 54.9 | 57.4 |