Performance evaluation of the proposed RBF SVM. Ten different training/testing splits are used. The mean and standard deviation (SD) of the metrics are shown. TPR denotes true positive rate; FPR denotes false positive rate; PPV denotes positive predictive value
| Algorithm | Proposed RBF SVM | |||
|---|---|---|---|---|
| Feature Set | #1 | #2 | #3 | |
| Accuracy | Mean | 98.0% | 94.6% | 98.4% |
| SD | 1.0% | 1.1% | 0.5% | |
| TPR (recall) | Mean | 91.5% | 86.5% | 93.8% |
| SD | 3.8% | 4.7% | 3.6% | |
| FPR | Mean | 0.7% | 3.6% | 0.7% |
| SD | 0.8% | 1.4% | 0.8% | |
| PPV (precision) | Mean | 96.8% | 83.8% | 97.1% |
| SD | 4.1% | 5.8% | 3.6% | |
| F1 score | Mean | 94% | 84.9% | 95.3% |
| SD | 2.8% | 2.7% | 1.6% | |