Sentiment Analysis On Semeval 2014 Task 4
Métriques
F1
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | F1 |
|---|---|
| exploiting-bert-for-end-to-end-aspect-based | 61.12 |
| grace-gradient-harmonized-and-cascaded | 70.71 |
| doer-dual-cross-shared-rnn-for-aspect-term | 60.35 |
| open-domain-targeted-sentiment-analysis-via | 68.06 |
| instructabsa-instruction-learning-for-aspect | 79.34 |
| an-interactive-multi-task-learning-network | 58.37 |
| a-unified-model-for-opinion-target-extraction | 57.9 |
| relation-aware-collaborative-learning-for | 63.4 |