"The highlighted tokens are predominantly Spanish (and some Portuguese) word endings and morphemes, such as verb and noun suffixes (-ar, -ado, -ción, -ión, -er, -os, -es, -ía, -ando, -iendo), as well as common function words (el, la, de, en, y, su, los, las). These patterns reflect grammatical inflections, gender/number agreement, and frequent connectors in Romance languages."
Score Type | Accuracy | Precision | Recall | F1 score | TPR | TNR | FPR | FNR |
---|---|---|---|---|---|---|---|---|
detection | 0.67 | 0.758 | 0.5 | 0.602 | 0.5 | 0.84 | 0.16 | 0.5 |
fuzz | 0.74 | 0.8 | 0.64 | 0.711 | 0.64 | 0.84 | 0.16 | 0.36 |