"The highlighted tokens are primarily morphemes, syllables, or short word fragments from multiple languages, especially Thai, Russian, and Bulgarian, often marking the start or core of content words such as nouns, verbs, or adjectives. These fragments frequently appear at the beginning or within important words, indicating their role in word formation and semantic content across diverse scripts and languages."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.82 | 0.864 | 0.76 | 0.809 | 0.76 | 0.88 | 0.12 | 0.24 |
fuzz | 0.58 | 0.544 | 0.98 | 0.7 | 0.98 | 0.18 | 0.82 | 0.02 |