Table 2.
Average random forest classification runtimes for predicting antibiotic vs. no-antibiotic treated neonatal microbiomes using read sampling intervals and concept drift (probability threshold = 0.9, k = 7, s = 42, decay ratio = 0.02, p = 8)
| Sampling interval (no. reads) | Runtime to reach initial classification (seconds) | Initial classification probability | Runtime to reach ≥ 0.9 classification probability (seconds) | Sampling intervals to reach ≥ 0.9 classification probability |
|---|---|---|---|---|
| No interval | 28.38 | 0.99 | 28.38 | na |
| 1,000,000 | 9.16 | 0.96 | 9.16 | 1 |
| 100,000 | 2.08 | 0.87 | 2.09 | 2 |
| 10,000 | 1.91 | 0.82 | 2.17 | 4 |
na not applicable