"The highlighted tokens are predominantly Vietnamese morphemes, words, or word parts that serve as key semantic units, including nouns, verbs, adjectives, and function words. These tokens often represent core concepts, actions, or attributes within a sentence, and include both standalone words and meaningful syllabic components in compound or multi-word expressions. The pattern reflects the analytic, syllable-based structure of Vietnamese, where individual morphemes carry significant meaning and are frequently combined to form compound words or phrases."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.92 | 1.0 | 0.84 | 0.913 | 0.84 | 1.0 | 0.0 | 0.16 |
fuzz | 0.93 | 0.978 | 0.88 | 0.926 | 0.88 | 0.98 | 0.02 | 0.12 |