Table 7 Results for deep learning.
From: Multilingual hope speech detection from tweets using transfer learning models
Language | Embedding’s | Models | Precision | Recall | F1-score | Accuracy |
|---|---|---|---|---|---|---|
English | FasText | CNN | 0.76 | 0.76 | 0.76 | 0.76 |
BiLSTM | 0.81 | 0.81 | 0.81 | 0.81 | ||
GloVe | CNN | 0.77 | 0.77 | 0.77 | 0.77 | |
BiLSTM | 0.8 | 0.79 | 0.79 | 0.79 | ||
Urdu | FasText | CNN | 0.71 | 0.51 | 0.35 | 0.51 |
BiLSTM | 0.75 | 0.75 | 0.74 | 0.75 | ||
GloVe | CNN | 0.54 | 0.51 | 0.37 | 0.51 | |
BiLSTM | 0.5 | 0.5 | 0.34 | 0.5 |
