"The highlighted tokens are primarily Japanese function words, particles, and common morphemes, as well as frequent kanji and katakana components found in names, titles, and grammatical constructions. There is a strong emphasis on elements that structure sentences, indicate relationships, or form part of proper nouns and set phrases, reflecting the syntactic and morphological backbone of Japanese text."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.81 | 1.0 | 0.62 | 0.765 | 0.62 | 1.0 | 0.0 | 0.38 |
fuzz | 0.8 | 0.917 | 0.66 | 0.767 | 0.66 | 0.94 | 0.06 | 0.34 |