Research Article
A Deep-Learning Framework for Analysing Students’ Review in Higher Education
Table 5
Precision (P), recall (R), and F1-score (F1) for aspects-term extraction models.
| | LSTM-CRF | GRU-CRF | Bi-LSTM-CRF | P | R | F1 | P | R | F1 | P | R | F1 |
| B-aspect | 0.69 | 0.69 | 0.69 | 0.61 | 0.70 | 0.65 | 0.74 | 0.74 | 0.74 | I-aspect | 0.66 | 0.24 | 0.35 | 0.62 | 0.38 | 0.47 | 0.62 | 0.46 | 0.53 | Micro avg | 0.69 | 0.53 | 0.60 | 0.61 | 0.58 | 0.60 | 0.70 | 0.64 | 0.67 | Macro avg | 0.68 | 0.53 | 0.57 | 0.61 | 0.58 | 0.59 | 0.70 | 0.64 | 0.66 |
|
|