"The highlighted tokens are predominantly morphemes, roots, or affixes within words in Slavic and some Turkic languages, often marking grammatical features such as case, number, gender, tense, or forming nouns and adjectives. These segments frequently appear at word boundaries or as part of compound or derived words, reflecting the agglutinative and inflectional nature of these languages."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.87 | 0.911 | 0.82 | 0.863 | 0.82 | 0.92 | 0.08 | 0.18 |
fuzz | 0.71 | 0.657 | 0.88 | 0.752 | 0.88 | 0.54 | 0.46 | 0.12 |