"The highlighted tokens are primarily morphemes, suffixes, or short word fragments in Portuguese (and some in other languages), often marking grammatical features such as gender, number, tense, or forming part of proper nouns and common words. Many are accented characters or endings that are important for word formation and meaning in the context of Romance languages."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.69 | 0.757 | 0.56 | 0.644 | 0.56 | 0.82 | 0.18 | 0.44 |
fuzz | 0.67 | 0.631 | 0.82 | 0.713 | 0.82 | 0.52 | 0.48 | 0.18 |