"The highlighted tokens are primarily morphemes, syllables, or word segments that are common in Hindi and English loanwords, especially those forming or ending nouns, adjectives, or technical terms. These segments often appear in compound words, transliterations, or borrowed terminology, and are frequently found in formal, technical, or institutional contexts."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.7 | 0.794 | 0.54 | 0.643 | 0.54 | 0.86 | 0.14 | 0.46 |
fuzz | 0.74 | 0.714 | 0.8 | 0.755 | 0.8 | 0.68 | 0.32 | 0.2 |