Table 8 Comparative experiments of different prompting methods on Restaurant dataset.
From: Graph-enhanced implicit aspect-level sentiment analysis based on multi-prompt fusion
Method | P | R | F1 |
|---|---|---|---|
T5 | 61.27 | 59.06 | 60.14 |
Fixed-sequence prompt | 61.38 | 62.07 | 61.72 |
Multi-prompt fusion | 63.69 | 62.63 | 63.15 |
