Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification

Limited approaches have been applied to Arabic sentiment analysis for a five-point classification problem. These approaches are based on single task learning with a handcrafted feature, which does not provide robust sentence representation. Recently, hierarchical attention networks have performed ou...

Full description

Bibliographic Details
Main Authors: Alali, Muath, Mohd Sharef, Nurfadhlina, Azmi Murad, Masrah Azrifah, Hamdan, Hazlina, Husin, Nor Azura
Format: Article
Published: MDPI 2021
Online Access:http://psasir.upm.edu.my/id/eprint/94301/
_version_ 1848861960376942592
author Alali, Muath
Mohd Sharef, Nurfadhlina
Azmi Murad, Masrah Azrifah
Hamdan, Hazlina
Husin, Nor Azura
author_facet Alali, Muath
Mohd Sharef, Nurfadhlina
Azmi Murad, Masrah Azrifah
Hamdan, Hazlina
Husin, Nor Azura
author_sort Alali, Muath
building UPM Institutional Repository
collection Online Access
description Limited approaches have been applied to Arabic sentiment analysis for a five-point classification problem. These approaches are based on single task learning with a handcrafted feature, which does not provide robust sentence representation. Recently, hierarchical attention networks have performed outstandingly well. However, when training such models as single-task learning, these models do not exhibit superior performance and robust latent feature representation in the case of a small amount of data, specifically on the Arabic language, which is considered a low-resource language. Moreover, these models are based on single task learning and do not consider the related tasks, such as ternary and binary tasks (cross-task transfer). Centered on these shortcomings, we regard five ternary tasks as relative. We propose a multitask learning model based on hierarchical attention network (MTLHAN) to learn the best sentence representation and model generalization, with shared word encoder and attention network across both tasks, by training three-polarity and five-polarity Arabic sentiment analysis tasks alternately and jointly. Experimental results showed outstanding performance of the proposed model, with high accuracy of 83.98%, 87.68%, and 84.59 on LABR, HARD, and BRAD datasets, respectively, and a minimum macro mean absolute error of 0.632% on the Arabic tweets dataset for five-point Arabic sentiment classification problem.
first_indexed 2025-11-15T13:09:26Z
format Article
id upm-94301
institution Universiti Putra Malaysia
institution_category Local University
last_indexed 2025-11-15T13:09:26Z
publishDate 2021
publisher MDPI
recordtype eprints
repository_type Digital Repository
spelling upm-943012023-05-08T02:50:18Z http://psasir.upm.edu.my/id/eprint/94301/ Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification Alali, Muath Mohd Sharef, Nurfadhlina Azmi Murad, Masrah Azrifah Hamdan, Hazlina Husin, Nor Azura Limited approaches have been applied to Arabic sentiment analysis for a five-point classification problem. These approaches are based on single task learning with a handcrafted feature, which does not provide robust sentence representation. Recently, hierarchical attention networks have performed outstandingly well. However, when training such models as single-task learning, these models do not exhibit superior performance and robust latent feature representation in the case of a small amount of data, specifically on the Arabic language, which is considered a low-resource language. Moreover, these models are based on single task learning and do not consider the related tasks, such as ternary and binary tasks (cross-task transfer). Centered on these shortcomings, we regard five ternary tasks as relative. We propose a multitask learning model based on hierarchical attention network (MTLHAN) to learn the best sentence representation and model generalization, with shared word encoder and attention network across both tasks, by training three-polarity and five-polarity Arabic sentiment analysis tasks alternately and jointly. Experimental results showed outstanding performance of the proposed model, with high accuracy of 83.98%, 87.68%, and 84.59 on LABR, HARD, and BRAD datasets, respectively, and a minimum macro mean absolute error of 0.632% on the Arabic tweets dataset for five-point Arabic sentiment classification problem. MDPI 2021 Article PeerReviewed Alali, Muath and Mohd Sharef, Nurfadhlina and Azmi Murad, Masrah Azrifah and Hamdan, Hazlina and Husin, Nor Azura (2021) Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification. Electronics, 11 (8). art. no. 1193. pp. 1-23. ISSN 2079-9292 https://www.mdpi.com/2079-9292/11/8/1193 10.3390/electronics11081193
spellingShingle Alali, Muath
Mohd Sharef, Nurfadhlina
Azmi Murad, Masrah Azrifah
Hamdan, Hazlina
Husin, Nor Azura
Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification
title Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification
title_full Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification
title_fullStr Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification
title_full_unstemmed Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification
title_short Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification
title_sort multitasking learning model based on hierarchical attention network for arabic sentiment analysis classification
url http://psasir.upm.edu.my/id/eprint/94301/
http://psasir.upm.edu.my/id/eprint/94301/
http://psasir.upm.edu.my/id/eprint/94301/