The valence analysis of speakers’ utterances or written posts helps to understand the activation and variations of the emotional state throughout the conversation. More recently, the concept of Emotion Carriers (EC) has been introduced to explain the emotion felt by the speaker and its manifestations. In this work, we investigate the natural inter-dependency of valence and ECs via a multi-task learning approach. We experiment with Pre-trained Language Models (PLM) for single-task, two-step, and joint settings for the valence and EC prediction tasks. We compare and evaluate the performance of generative (GPT-2) and discriminative (BERT) architectures in each setting. We observed that providing the ground truth label of one task improves the prediction performance of the models in the other task. We further observed that the discriminative model achieves the best trade-off of valence and EC prediction tasks in the joint prediction setting. As a result, we attain a single model that performs both tasks, thus, saving computation resources at training and inference times.

Understanding Emotion Valence is a Joint Deep Learning Task / Roccabruna, Gabriel; Mousavi, Seyed Mahed; Riccardi, Giuseppe. - (2023), pp. 85-95. (Intervento presentato al convegno ACL tenutosi a Toronto, Canada nel 9th-14th July 2023).

Understanding Emotion Valence is a Joint Deep Learning Task

Gabriel Roccabruna
Primo
;
Seyed Mahed Mousavi
Secondo
;
Giuseppe Riccardi
Ultimo
2023-01-01

Abstract

The valence analysis of speakers’ utterances or written posts helps to understand the activation and variations of the emotional state throughout the conversation. More recently, the concept of Emotion Carriers (EC) has been introduced to explain the emotion felt by the speaker and its manifestations. In this work, we investigate the natural inter-dependency of valence and ECs via a multi-task learning approach. We experiment with Pre-trained Language Models (PLM) for single-task, two-step, and joint settings for the valence and EC prediction tasks. We compare and evaluate the performance of generative (GPT-2) and discriminative (BERT) architectures in each setting. We observed that providing the ground truth label of one task improves the prediction performance of the models in the other task. We further observed that the discriminative model achieves the best trade-off of valence and EC prediction tasks in the joint prediction setting. As a result, we attain a single model that performs both tasks, thus, saving computation resources at training and inference times.
2023
Proceedings of the 13th Workshop on Computational Approaches to Subjectivity, Sentiment, & Social Media Analysis
Toronto, Canada
Association for Computational Linguistics
978-1-959429-87-6
Roccabruna, Gabriel; Mousavi, Seyed Mahed; Riccardi, Giuseppe
Understanding Emotion Valence is a Joint Deep Learning Task / Roccabruna, Gabriel; Mousavi, Seyed Mahed; Riccardi, Giuseppe. - (2023), pp. 85-95. (Intervento presentato al convegno ACL tenutosi a Toronto, Canada nel 9th-14th July 2023).
File in questo prodotto:
File Dimensione Formato  
2023.wassa-1.9.pdf

accesso aperto

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Creative commons
Dimensione 507.55 kB
Formato Adobe PDF
507.55 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/391950
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact