"The highlighted tokens consistently correspond to the morphemes or substrings forming the country name \"Brazil\" and its derivatives across multiple languages and scripts, often marking the root or core segment of the word regardless of linguistic context."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.56 | 0.875 | 0.14 | 0.241 | 0.14 | 0.98 | 0.02 | 0.86 |
fuzz | 0.6 | 1.0 | 0.2 | 0.333 | 0.2 | 1.0 | 0.0 | 0.8 |