Classical Machine Learning and Large Models for Text-Based Emotion Recognition

In the era of digital communication, detecting emotions has become crucial for applications ranging from customer service to mental health assessment. This study examines emotion recognition in text through various machine learning techniques, from traditional machine learning techniques, to advance...

Full description

Saved in:
Bibliographic Details
Published in:Procedia computer science 2024, Vol.241, p.77-84
Main Authors: Esfahani, Seyed Hamed Noktehdan, Adda, Mehdi
Format: Article
Language:English
Subjects:
Emotion Recognition
Fine-tuning
Large Language Models
LLM
Machine Learning
NLP
One-shot learning
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the era of digital communication, detecting emotions has become crucial for applications ranging from customer service to mental health assessment. This study examines emotion recognition in text through various machine learning techniques, from traditional machine learning techniques, to advanced large language models (LLMs) such as BERT, Falcon 7B, and Mistral 7B. The results reveal that the fine-tuned Mistral 7B model is the most precise, achieving an accuracy of 76%. Furthermore, Support Vector Machine attained an accuracy of 64%.
ISSN:1877-0509
1877-0509
DOI:10.1016/j.procs.2024.08.013