Performance evaluation of the random forest using Feature Set #3. Ten different training/testing splits are used. The mean and standard deviation (SD) of the metrics are shown. FPR, TPR, and PPV denote false positive rate, true positive rate, and positive predictive value, respectively.
| Random Forest | |||
|---|---|---|---|
| Algorithm | Dual | Triple | |
| Accuracy | Mean | 98.4% | 99.0% |
| SD | 0.4% | 0.5% | |
| FPR (False Alarm) | Mean | 0.5% | 0.4% |
| SD | 0.3% | 0.2% | |
| TPR (Recall) | Mean | 95.5% | 97.6% |
| SD | 2.4% | 2.9% | |
| PPV (Precision) | Mean | 97.7% | 98.0% |
| SD | 1.4% | 0.9% | |
| F1 score | Mean | 96.5% | 97.8% |
| SD | 1.2% | 1.6% | |