Table 9 The performance of different classification algorithms on 2-class-dataset according to fGOS. Significant values are in bold.
From: Prognosis prediction in traumatic brain injury patients using machine learning algorithms
Algorithm | Acc (%) | Acc Rank | Prec (%) | Rec (%) | AUC |
|---|---|---|---|---|---|
NB | 78.65 ± 3.93 | 7 | 55.95 ± 8.59 | 53.82 ± 8.36 | 0.812 ± 0.039 |
RF | 80.88 ± 1.86 | 3 | 77.32 ± 10.90 | 29.04 ± 4.50 | 0.807 ± 0.035 |
KNN(k = 5) | 76.95 ± 1.66 | 9 | 54.49 ± 7.76 | 28.54 ± 2.95 | 0.661 ± 0.060 |
KNN(k = 6) | 76.95 ± 1.96 | 10 | 53.56 ± 7.68 | 29.27 ± 7.31 | 0.675 ± 0.052 |
DT | 78.22 ± 1.23 | 8 | 63.26 ± 9.17 | 25.33 ± 8.72 | 0.683 ± 0.039 |
RI | 80.70 ± 2.51 | 4 | 66.69 ± 9.83 | 41.18 ± 7.83 | 0.758 ± 0.054 |
DL | 78.95 ± 2.74 | 6 | 55.82 ± 5.73 | 59.37 ± 10.79 | 0.821 ± 0.038 |
GBT | 79.25 ± 3.02 | 5 | 57.51 ± 6.86 | 56.56 ± 7.68 | 0.823 ± 0.015 |
LR | 81.61 ± 2.58 | 2 | 67.61 ± 8.10 | 45.99 ± 4.21 | 0.834 ± 0.031 |
GLM | 82.03 ± 2.34 | 1 | 68.00 ± 6.36 | 47.22 ± 8.41 | 0.834 ± 0.038 |
Avg Acc | 79.42 |
