Selective Reading for Arabic Sentiment Analysis

This work introduces a novel deep learning method for Arabic sentiment analysis, arguing that reading the entire input sequence is not always necessary. Many texts can be accurately classified without processing all input tokens. The method employs a reinforcement learning agent that selects relevan...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohamed Zouidine, Mohammed Khalil
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10947753/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849764234285547520
author Mohamed Zouidine
Mohammed Khalil
author_facet Mohamed Zouidine
Mohammed Khalil
author_sort Mohamed Zouidine
collection DOAJ
description This work introduces a novel deep learning method for Arabic sentiment analysis, arguing that reading the entire input sequence is not always necessary. Many texts can be accurately classified without processing all input tokens. The method employs a reinforcement learning agent that selects relevant tokens using a selection policy network. Instead of predicting sentiment polarity from the entire input, the model focuses only on tokens chosen by the policy network. To empirically evaluate the proposed method, experiments were carried out on three Arabic sentiment analysis datasets: Large Arabic Book Reviews (LABR), Hotels Arabic Reviews Data (HARD), and Arabic Sentiment Tweets Dataset (ASTD). The results demonstrate a significant improvement in Arabic sentiment classification with the selective reading method, achieving state-of-the-art accuracy while using only a fraction of the tokens. However, the approach introduces additional computational cost due to the reinforcement learning component, and its scalability to larger datasets might require further optimization.
format Article
id doaj-art-d1c6fb52adfa4e8a8537a9e2df207342
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d1c6fb52adfa4e8a8537a9e2df2073422025-08-20T03:05:11ZengIEEEIEEE Access2169-35362025-01-0113591575916910.1109/ACCESS.2025.355697610947753Selective Reading for Arabic Sentiment AnalysisMohamed Zouidine0https://orcid.org/0000-0002-0848-1027Mohammed Khalil1LMCSA, FSTM, Hassan II University of Casablanca, Casablanca, MoroccoLMCSA, FSTM, Hassan II University of Casablanca, Casablanca, MoroccoThis work introduces a novel deep learning method for Arabic sentiment analysis, arguing that reading the entire input sequence is not always necessary. Many texts can be accurately classified without processing all input tokens. The method employs a reinforcement learning agent that selects relevant tokens using a selection policy network. Instead of predicting sentiment polarity from the entire input, the model focuses only on tokens chosen by the policy network. To empirically evaluate the proposed method, experiments were carried out on three Arabic sentiment analysis datasets: Large Arabic Book Reviews (LABR), Hotels Arabic Reviews Data (HARD), and Arabic Sentiment Tweets Dataset (ASTD). The results demonstrate a significant improvement in Arabic sentiment classification with the selective reading method, achieving state-of-the-art accuracy while using only a fraction of the tokens. However, the approach introduces additional computational cost due to the reinforcement learning component, and its scalability to larger datasets might require further optimization.https://ieeexplore.ieee.org/document/10947753/Arabic natural language processingconvolutional neural networkdeep reinforcement learninggated recurrent unitselective reading
spellingShingle Mohamed Zouidine
Mohammed Khalil
Selective Reading for Arabic Sentiment Analysis
IEEE Access
Arabic natural language processing
convolutional neural network
deep reinforcement learning
gated recurrent unit
selective reading
title Selective Reading for Arabic Sentiment Analysis
title_full Selective Reading for Arabic Sentiment Analysis
title_fullStr Selective Reading for Arabic Sentiment Analysis
title_full_unstemmed Selective Reading for Arabic Sentiment Analysis
title_short Selective Reading for Arabic Sentiment Analysis
title_sort selective reading for arabic sentiment analysis
topic Arabic natural language processing
convolutional neural network
deep reinforcement learning
gated recurrent unit
selective reading
url https://ieeexplore.ieee.org/document/10947753/
work_keys_str_mv AT mohamedzouidine selectivereadingforarabicsentimentanalysis
AT mohammedkhalil selectivereadingforarabicsentimentanalysis