SpanEmo: Casting Multi-label Emotion Classification as Span-prediction
About
Emotion recognition (ER) is an important task in Natural Language Processing (NLP), due to its high impact in real-world applications from health and well-being to author profiling, consumer analysis and security. Current approaches to ER, mainly classify emotions independently without considering that emotions can co-exist. Such approaches overlook potential ambiguities, in which multiple emotions overlap. We propose a new model "SpanEmo" casting multi-label emotion classification as span-prediction, which can aid ER models to learn associations between labels and words in a sentence. Furthermore, we introduce a loss function focused on modelling multiple co-existing emotions in the input sentence. Experiments performed on the SemEval2018 multi-label emotion data over three language sets (i.e., English, Arabic and Spanish) demonstrate our method's effectiveness. Finally, we present different analyses that illustrate the benefits of our method in terms of improving the model performance and learning meaningful associations between emotion classes and words in the sentence.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Multi-label emotion classification | SemEval Task 1:E-c 2018 (test) | Macro F157.8 | 53 | |
| Emotion Classification | SemEval Task 1 Spanish 2018 (test) | Micro F164.1 | 11 | |
| Multi-label Text Classification | Indonesian dataset | Macro F171.62 | 11 | |
| Emotion Classification | SemEval Task 1 English 2018 (test) | Macro F157.8 | 10 | |
| Multi-label emotion classification | Arabic dataset (test) | F1-Macro53.63 | 10 | |
| Multi-label Text Classification | Spanish dataset | F-Macro55.49 | 10 |