Multilingual, monolingual and mono-dialectal transfer learning for Moroccan Arabic sentiment classification

Transfer learning has recently proven to be very powerful in diverse natural language processing (NLP) tasks such as Machine translation, Sentiment Analysis, and Question/Answering. In this work, we investigate the use of transfer learning (TL) in Dialectal Arabic sentiment classification. Our main...

Full description

Saved in:
Bibliographic Details
Published in:Social network analysis and mining 2023-12, Vol.14 (1), p.3
Main Authors: Boudad, Naaima, Faizi, Rdouan, Oulad Haj Thami, Rachid
Format: Article
Language:English
Subjects:
Arabic language
Bidirectionality
Classification
Code switching
Data mining
Dialects
Hate speech
Knowledge management
Language modeling
Learning
Machine translation
Natural language processing
Sentiment analysis
Translation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Transfer learning has recently proven to be very powerful in diverse natural language processing (NLP) tasks such as Machine translation, Sentiment Analysis, and Question/Answering. In this work, we investigate the use of transfer learning (TL) in Dialectal Arabic sentiment classification. Our main objective is to enhance the performance of Sentiment classification and overcome the low resource issue of Arabic dialect. To this end, we use Bidirectional Encoder Representation from Transformers (BERT) to transfer contextual knowledge learned from language modeling task to sentiment classification. We particularly use the multilingual models mBert and XLM-Roberta, the specific Arabic models ARABERT, MARBERT, QARIB, CAMEL and the specific Moroccan dialect DarijaBert. After carrying out downstream fine-tuning experiments using different Moroccan SA datasets, we found that using TL significantly increases the performance of sentiment classification in Moroccan Arabic. Nevertheless, though specific Arabic models have proven to perform much better than multilingual and dialectal models, our experiments have demonstrated that multilingual models can be more effective in texts characterized by an extensive use of code-switching.
ISSN:1869-5450
1869-5469
DOI:10.1007/s13278-023-01159-9