Research Article
PCG Classification Using Multidomain Features and SVM Classifier
Table 14
Classification performance based on features in specified domain.
| Rank | Domain (# features) | Mean of overall score | Standard deviation |
| 1 | Frequency spectrum of state (308) | 0.85 | 0.021 | 2 | Entropy (16) | 0.82 | 0.028 | 3 | Energy (47) | 0.78 | 0.020 | 4 | Cepstrum (65) | 0.75 | 0.027 | 5 | High-order statistics (16) | 0.73 | 0.029 | 6 | Frequency spectrum of records (27) | 0.71 | 0.025 | 7 | Time interval (20) | 0.70 | 0.025 | 8 | Cyclostationarity (4) | 0.65 | 0.042 | 9 | State amplitude (12) | 0.61 | 0.025 |
|
|