"The highlighted tokens are primarily function words, common morphemes, and noun or verb roots in Hindi and related scripts, often marking grammatical relationships, inflections, or forming the core of named entities and key content words. These tokens are essential for sentence structure, meaning, and the identification of important entities or actions within the text."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.79 | 0.939 | 0.62 | 0.747 | 0.62 | 0.96 | 0.04 | 0.38 |
fuzz | 0.71 | 0.678 | 0.8 | 0.734 | 0.8 | 0.62 | 0.38 | 0.2 |