Emotion Recognition In Conversation On Emowoz
평가 지표
Macro F1
Macro F1 (w/o Neutral)
Weighted F1
Weighted F1 (w/o Neutral)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Macro F1 | Macro F1 (w/o Neutral) | Weighted F1 | Weighted F1 (w/o Neutral) | Paper Title | Repository |
|---|---|---|---|---|---|---|
| DialogueRNN-GloVe | 46.33 | 40.14 | 80.76 | 74.56 | EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems | |
| CD-ERC | 65.33 | - | - | - | Context-Dependent Embedding Utterance Representations for Emotion Recognition in Conversations | |
| DialogueRNN-BERT | 57.10 | 52.15 | 83.41 | 75.50 | EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems | |
| ContextBERT | 59.79 | 54.30 | 88.33 | 79.67 | EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems | |
| COSMIC | 61.12 | 56.34 | 85.94 | 77.09 | EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems | |
| BERT | 55.80 | 50.14 | 84.83 | 73.55 | EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems |
0 of 6 row(s) selected.