Exclusion method | Zero-cutoff | Max valid | Max test |
---|
| n | ACC (in %) | F1 (in %) | n | ACC (in %) | F1 (in %) | n | ACC (in %) | F1 (in %) |
---|
Random (-) | – | – | – | 41 | 58.57 | 66.42 | 445 | 66.07 | 75.81 |
LOO (LR) | 84 | 62.50 | 68.81 | 102 | 56.79 | 64.44 | 402 | 68.21 | 76.33 |
LOO (RF) | 49 | 57.86 | 65.40 | 23 | 61.79 | 69.45 | 186 | 68.93 | 76.63 |
Data Shapley (LR) | 152 | 61.79 | 69.52 | 340 | 61.07 | 65.79 | 399 | 73.57 | 75.74 |
Data Shapley (RF) | 164 | 57.86 | 68.94 | 134 | 60.36 | 70.05 | 14 | 65.36 | 72.30 |
- Different methods were used to identify and focus on the training subjects with the most informative data. The zero-cutoff method excluded all training subjects with Data Shapley values smaller than zero. Max valid was the threshold achieved by maximizing the results for the independent validation set. Max test was the optimistic threshold which achieved the best results for the test set. Ten repetitions with different seeds were performed for every exclusion data set. The best results are highlighted in bold