Table 2.
Linguistic Scale | Triangular Fuzzy Scale | Criteria |
---|---|---|
Very High (VH) | (0.75, 1, 1) | Accuracy, recall, precision F1-score, ROC-AUC, log loss |
High (H) | (0.50, 0.75, 1) | Number of training samples needed, the impact of feature scaling, the impact of hyper-parameter tuning |
Medium (M) | (0.25, 0.50, 0.75) | Tolerance to irrelevant attributes |
Low (L) | (0, 0.25, 0.50) | - |
Very Low (VL) | (0, 0, 0.25) | - |