Table 9 Class wise score of BERT model in term of English and Urdu dataset.
From: Multilingual hope speech detection from tweets using transfer learning models
Language | Class | Precision | Recall | F1-score | Support | Accuracy |
|---|---|---|---|---|---|---|
English | Not hope | 0.88 | 0.84 | 0.86 | 922 | 0.87 |
Hope | 0.85 | 0.89 | 0.87 | 932 | ||
Urdu | Not hope | 0.74 | 0.88 | 0.8 | 926 | 0.79 |
Hope | 0.85 | 0.69 | 0.76 | 928 |
