"The highlighted tokens are primarily Vietnamese morphemes, syllables, or word segments, often marking the start of words or compounds, and include both native and Sino-Vietnamese roots. Many are high-frequency function words, affixes, or common noun/adjective stems, and several are associated with grammatical or semantic roles such as denoting people, places, actions, or qualities. There is a notable emphasis on tokens with the \"ê\"/\"ệ\"/\"ế\"/\"ề\" vowel, as well as on tokens that form part of compound nouns, proper names, or technical terms."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.66 | 0.9 | 0.36 | 0.514 | 0.36 | 0.96 | 0.04 | 0.64 |
fuzz | 0.74 | 1.0 | 0.48 | 0.649 | 0.48 | 1.0 | 0.0 | 0.52 |