ArCAR: A Novel Deep Learning Computer-Aided Recognition for Character-Level Arabic Text Representation and Recognition

Abdullah Y Muaad,Sungyoung Lee,Hanumanthappa Jayappa,Mugahed A Al-Antari

doi:10.3390/a14070216

Abdullah Y Muaad, Sungyoung Lee + Show 2 more

Open Access

https://doi.org/10.3390/a14070216

Copy DOI

Abstract

Arabic text classification is a process to simultaneously categorize the different contextual Arabic contents into a proper category. In this paper, a novel deep learning Arabic text computer-aided recognition (ArCAR) is proposed to represent and recognize Arabic text at the character level. The input Arabic text is quantized in the form of 1D vectors for each Arabic character to represent a 2D array for the ArCAR system. The ArCAR system is validated over 5-fold cross-validation tests for two applications: Arabic text document classification and Arabic sentiment analysis. For document classification, the ArCAR system achieves the best performance using the Alarabiya-balance dataset in terms of overall accuracy, recall, precision, and F1-score by 97.76%, 94.08%, 94.16%, and 94.09%, respectively. Meanwhile, the ArCAR performs well for Arabic sentiment analysis, achieving the best performance using the hotel Arabic reviews dataset (HARD) balance dataset in terms of overall accuracy and F1-score by 93.58% and 93.23%, respectively. The proposed ArCAR seems to provide a practical solution for accurate Arabic text representation, understanding, and classification.

Highlights

The evaluation results of the proposed Arabic text computer-aided recognition (ArCAR) system for Arabic text recognition are presented for both applications of documents classification and sentiment analysis
Different datasets with multiple classes were used to test the reliability of the proposed ArCAR system
For Arabic sentiment analysis, the number of posts or comments varies between 165 k and 500 k, while the dataset has two and three balance and unbalance classes, respectively

Summary

Introduction

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Natural language processing (NLP) is a subfield of linguistics, computer science, and artificial intelligence concerned with computers and human language interactions. The goal of NLP is to process textual contents and extract the most useful information for effective real-life decisions. Text mining problems have gained much attention and have become a vital research area because of the boom in textual applications such as document recognition, social networking gates, or text identification from images or paintings [1]

Objectives

Methods

Results

Discussion

Conclusion