WASSA@IITK at WASSA 2021: Multi-task Learning and Transformer Finetuning for Emotion Classification and Empathy Prediction
This work addresses emotion and empathy modeling for natural language processing applications, but it is incremental as it applies existing methods like ELECTRA and multi-task learning to a specific shared task.
The paper tackled emotion classification and empathy prediction from essays using ELECTRA and multi-task learning, achieving a Pearson Correlation Coefficient of 0.533 for empathy prediction and a macro F1 score of 0.5528 for emotion classification, ranking 1st and 3rd in the respective WASSA 2021 shared tasks.
This paper describes our contribution to the WASSA 2021 shared task on Empathy Prediction and Emotion Classification. The broad goal of this task was to model an empathy score, a distress score and the overall level of emotion of an essay written in response to a newspaper article associated with harm to someone. We have used the ELECTRA model abundantly and also advanced deep learning approaches like multi-task learning. Additionally, we also leveraged standard machine learning techniques like ensembling. Our system achieves a Pearson Correlation Coefficient of 0.533 on sub-task I and a macro F1 score of 0.5528 on sub-task II. We ranked 1st in Emotion Classification sub-task and 3rd in Empathy Prediction sub-task