Table 2 Evaluation results for codes having a probability greater than 0.5.
From: Multi-label text classification via secondary use of large clinical real-world data sets
Model | Precision | Recall | F1-measure | MAP |
|---|---|---|---|---|
medBERT.de | 0.788 | 0.753 | 0.756 | 0.809 |
surgeryBERT.at | 0.773 | 0.729 | 0.736 | 0.792 |
fastText | 0.726 | 0.708 | 0.700 | 0.763 |
CNN | 0.710 | 0.661 | 0.670 | 0.730 |
SVM | 0.759 | 0.717 | 0.723 | 0.784 |
LR | 0.596 | 0.557 | 0.562 | 0.614 |
