"The highlighted tokens are predominantly suffixes, inflections, and morphemes in Turkish and related languages, as well as proper nouns, dates, and grammatical particles. These tokens often mark case, possession, plurality, tense, or are part of compound words and names, reflecting the agglutinative structure of the language and the importance of morphological boundaries and named entities in text processing."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.94 | 0.923 | 0.96 | 0.941 | 0.96 | 0.92 | 0.08 | 0.04 |
fuzz | 0.91 | 0.887 | 0.94 | 0.913 | 0.94 | 0.88 | 0.12 | 0.06 |