"The highlighted tokens are often single characters, syllables, or short morphemes from multiple languages, especially Thai, as well as fragments of words in other languages. These tokens tend to be linguistically meaningful units such as prefixes, suffixes, or roots, and are frequently found at the beginning, middle, or end of words, often marking grammatical or semantic boundaries."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.62 | 0.573 | 0.94 | 0.712 | 0.94 | 0.3 | 0.7 | 0.06 |
fuzz | 0.56 | 0.532 | 1.0 | 0.694 | 1.0 | 0.12 | 0.88 | 0.0 |