Table 15 Class-wise scores using best-performing models for Spanish (xlm-roberta-base) and German (bert-base-german-dbmdz-uncased).
From: Multilingual identification of nuanced dimensions of hope speech in social media texts
Language | Class | Precision | Recall | F1-score |
|---|---|---|---|---|
German | Binary | |||
Not Hope | 0.9096 | 0.8587 | 0.8834 | |
Hope | 0.8226 | 0.8847 | 0.8525 | |
Multiclass | ||||
Not Hope | 0.8898 | 0.8845 | 0.8872 | |
Generalized Hope | 0.7920 | 0.7995 | 0.7958 | |
Realistic Hope | 0.5908 | 0.6282 | 0.6089 | |
Unrealistic Hope | 0.4807 | 0.4328 | 0.4555 | |
Spanish | Binary | |||
Not Hope | 0.8478 | 0.7965 | 0.8213 | |
Hope | 0.8006 | 0.8510 | 0.8250 | |
Multiclass | ||||
Not Hope | 0.8546 | 0.8145 | 0.8341 | |
Generalized Hope | 0.7189 | 0.7436 | 0.7311 | |
Realistic Hope | 0.5035 | 0.5632 | 0.5317 | |
Unrealistic Hope | 0.5920 | 0.6041 | 0.5980 | |
