"The highlighted tokens are primarily Korean morphemes, words, and grammatical endings, often marking key semantic or syntactic units such as nouns, verbs, particles, and endings that define sentence structure or meaning. There is also frequent highlighting of proper nouns, numbers, and technical terms, as well as some corrupted or non-standard characters, indicating a focus on linguistically significant elements and possibly on tokens that are informative for parsing or translation tasks."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.74 | 0.962 | 0.5 | 0.658 | 0.5 | 0.98 | 0.02 | 0.5 |
fuzz | 0.74 | 0.929 | 0.52 | 0.667 | 0.52 | 0.96 | 0.04 | 0.48 |