NOVEL MULTI-CHANNEL DEEP LEARNING MODEL FOR ARABIC NEWS CLASSIFICATION

(Received: 4-Jul.-2024, Revised: 11-Aug.-2024 and 26-Aug.-2024 , Accepted: 31-Aug.-2024)

Authors Imad Jamaleddyn, Rachid El Ayachi, Mohamed Biniz,

Keywords #Convolutional neural networks (CNNs) #Long Short-Term Memory (LSTM) #Gated Recurrent Units (GRUs) #Word embedding #Arabic-text classification

Abstract In the era of digital journalism, the classification of Arabic news presents a significant challenge due to the complex nature of the language and the vast diversity of content. This study introduces a novel multi-channel deep-learning model, CLGNet, designed to enhance the accuracy of Arabic-news categorization. By integrating Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs), the proposed model effectively processes and classifies Arabic-text data. Extensive experiments were conducted on multiple datasets, including CNN, BBC and OSAC, where the model achieved outstanding accuracy and robustness, outperforming existing methods. The findings underscore the effectiveness of our hybrid model in addressing the challenges of Arabic-text classification and its potential applications in automated news categorization systems.

References

[1] Y. Timmerman and A. Bronselaer, "Measuring Data Quality in Information Systems Research,"Decision Support Systems, vol. 126, p. 113138, DOI: 10.1016/j.dss.2019.113138, Nov. 2019.

[2] C. Porlezza, "Accuracy in Journalism," Oxford Research Encyclopedia of Communication, DOI:10.1093/acrefore/9780190228613.013.773, Oxford University Press, Mar. 2019.

[3] N. Newman, R. Fletcher, A. Schulz, S. Andi, C. T. Robertson and R. K. Nielsen, Reuters InstituteDigital News Report 2021, Reuters Institute for the Study of Journalism, pp. 1-164, 10th Edn, [Online], Available:https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2021-06/Digital_News_Report_202
1_FINAL.pdf, 2021.

[4] I. Ahmad, F. AlQurashi and R. Mehmood, "Machine and Deep Learning Methods with Manual andAutomatic Labelling for News Classification in Bangla Language," arXiv: 2210.10903, DOI: 10.48550/arXiv.2210.10903, 2022.

[5] R. Indrakumari, T. Poongodi and K. Singh, "Introduction to Deep Learning," in Book: Advanced DeepLearning for Engineers, pp. 1–22, DOI: 10.1007/978-3-030-66519-7_1, Springer International Publishing, 2021.

[6] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser and I. Polosukhin,"Attention Is All You Need," Proc. of the 31st Conference on Neural Information Processing Systems (NIPS 2017), pp. 1-11, Long Beach, CA, USA, 2017.

[7] H. Hassan et al., "Achieving Human Parity on Automatic Chinese to English News Translation," arXiv:1803.05567, DOI: 10.48550/arXiv.1803.05567, 2018.

[8] M. Jabrane, I. Hafidi and Y. Rochd, "An Improved Active Machine Learning Query Strategy for EntityMatching Problem," Proc. of Advances in Machine Intelligence and Computer Science Applications (ICMICSA 2022), pp. 317–327, Springer Nature Switzerland, 2023.

[9] J. Mourad, T. Hiba, R. Yassir and H. Imad, "ERABQS: Entity Resolution Based on Active MachineLearning and Balancing Query Strategy," Journal of Intelligent Information Systems, DOI: 10.1007/s10844-024-00853-0, Mar. 2024.

[10] M. Jabrane, H. Tabbaa, A. Hadri and I. Hafidi, "Enhancing Entity Resolution with a Hybrid ActiveMachine Learning Framework: Strategies for Optimal Learning in Sparse Datasets," Information Systems, vol. 125, p. 102410, Nov. 2024.

[11] J. Devlin, M.-W. Chang, K. Lee and K. Toutanova, "Bert: Pre-training of Deep BidirectionalTransformers for Language Understanding," Proc. of NAACL-HLT 2019, pp. 4171–4186, Minneapolis, Minnesota, June 2 - June 7, 2019.

[12] X. Liu, P. He, W. Chen and J. Gao, "Multi-task Deep Neural Networks for Natural LanguageUnderstanding," Proc. of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 4487–4496, Florence, Italy, July 28 - August 2, 2019.

[13] P. Rajpurkar, J. Zhang, K. Lopyrev and P. Liang, "Squad: 100, 000+ Questions for MachineComprehension of Text," Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing, pp. 2383–2392, Austin, Texas, USA, 2016.

[14] G. Lample and A. Conneau, "Cross-lingual Language Model Pre-training," Proc. of the 33rd Conf. onNeural Information Processing Systems (NeurIPS 2019), pp. 1-11, Vancouver, Canada, 2019.

[15] K. M. Fouad, S. F. Sabbeh and W. Medhat, "Arabic Fake News Detection Using Deep Learning,"Computers, Materials & Continua, vol. 71, no. 2, pp. 3647–3665, 2022.

[16] M. Azzeh, A. Qusef and O. Alabboushi, "Arabic Fake News Detection in Social Media Context UsingWord Embeddings and Pre-trained Transformers," Arabian Journal for Science and Engineering, DOI: 10.1007/s13369-024-08959-x, Apr. 2024.

[17] M. M. Abdelsamie, S. S. Azab and H. A. Hefny, "A Comprehensive Review on Arabic OffensiveLanguage and Hate Speech Detection on Social Media: Methods, Challenges and Solutions," Social Network Analysis and Mining, vol. 14, p. 111, DOI: 10.1007/s13278-024-01258-1, May 2024.

[18] L. Zhang, W. Jiang and Z. Zhao, "Short-text Feature Expansion and Classification Based on Non- negative Matrix Factorization," Proc. of Machine Learning for Cyber Security(ML4CS 2020), pp. 347–362, DOI: 10.1007/978-3-030-62463-7_32, Springer International Publishing, 2020.

[19] M. S. H. Ameur, R. Belkebir and A. Guessoum, "Robust Arabic Text Categorization by CombiningConvolutional and Recurrent Neural Networks," ACM Transactions on Asian and Low-Resource Language Information Processing, vol. 19, no. 5, Article no. 66, July 2020.

[20] A. M. Bdeir and F. Ibrahim, "A Framework for Arabic Tweets Multi-label Classification Using WordEmbedding and Neural Networks Algorithms," Proc. of the 2020 2nd Int. Conf. on Big Data Engineering (BDE’ 2020), pp. 105-112, DOI: 10.1145/3404512.340452, ACM, May 2020.

[21] A. Hassanein and M. Nour, "A Proposed Model of Selecting Features for Classifying Arabic Text,"Jordanian J. of Computers and Information Technology (JJCIT), vol. 5, no. 3, pp. 275-290, Dec. 2019.

[22] L. Qadi, H. Rifai, S. Obaid and A. Elnagar, "A Scalable Shallow Learning Approach for TaggingArabic News Articles," Jordanian Journal of Computers and Information Technology (JJCIT), vol. 6, no. 3, pp. 263-280, 2020.

[23] T. A. Wotaifi and B. N. Dhannoon, "An Effective Hybrid Deep Neural Network for Arabic Fake NewsDetection," Baghdad Science Journal, vol. 20, no. 4, DOI: 10.21123/bsj.2023.7427, Jan. 2023.

[24] A. B. Nassif, A. Elnagar, O. Elgendy and Y. Afadar, "Arabic Fake News Detection Based on DeepContextualized Embedding Models," Neural Computing and Applications, vol. 34, pp. 16019–16032, May 2022.

[25] R. Romero, P. Celard, J. Sorribes-Fdez, A. Seara Vieira, E. Iglesias and L. Borrajo, "MobyDeep: ALightweight CNN Architecture to Configure Models for Text Classification," Knowledge-based Systems, vol. 257, p. 109914, DOI: 10.1016/j.knosys.2022.109914, Dec. 2022.

[26] A. Alqahtani, H. Ullah Khan, S. Alsubai, M. Sha, A. Almadhor, T. Iqbal and S. Abbas, "An EfficientApproach for Textual Data Classification Using Deep Learning," Frontiers in Computational Neuroscience, vol. 16, DOI: 10.3389/fncom.2022.992296, Sept. 2022.

[27] A. Awajan, "Arabic Text Pre-processing for the Natural Language Processing Applications," Arab GulfJournal of Scientific Research, vol. 25, no. 4, pp. 179–189, 2007.

[28] Y. Sun et al., "Modifying the One-hot Encoding Technique Can Enhance the Adversarial Robustness ofthe Visual Model for Symbol Recognition," Expert Systems with Applications, vol. 250, p. 123751, DOI: 10.1016/j.eswa.2024.123751, Sept. 2024.

[29] N. Alalyani and S. Larabi, "NADA: New Arabic Dataset for Text Classification," Int. Journal ofAdvanced Computer Science and Applications, vol. 9, no. 9, DOI: 10.14569/IJACSA.2018.090928, 2018.

[30] M. Hossin and M. N. Sulaiman, "A Review on Evaluation Metrics for Data Classification Evaluations,"Int. Journal of Data Mining & Knowledge Management Process, vol. 5, no. 2, pp. 01–11, DOI : 10.5121/ijdkp.2015.5201, 2015.

[31] S. Bahassine, A. Madani, M. Al-Sarem and M. Kissi, "Feature Selection Using an Improved Chi-squarefor Arabic Text Classification," Journal of King Saud University - Computer and Information Sciences, vol. 32, no. 2, pp. 225–231, Feb. 2020.

[32] I. Jamaleddyn and M. Biniz, "Contribution to Arabic Text Classification Using Machine LearningTechniques," Proc. of Business Intelligence (CBI 2021), pp. 18–32, DOI: 10.1007/978-3-030-76508-8_2, Springer, 2021.

[33] A. Y. Muaad, H. Jayappa, M. A. Al-antari and S. Lee, "ArCAR: A Novel Deep Learning Computer-aided Recognition for Character-level Arabic Text Representation and Recognition," Algorithms, vol. 14, no. 7, p. 216, DOI: 10.3390/a14070216, July 2021.

[34] T. Sabri, O. E. Beggar and M. Kissi, "Comparative Study of Arabic Text Classification Using FeatureVectorization Methods," Procedia Computer Science, vol. 198, pp. 269–275, DOI: 10.1016/j.procs.2021.12.239, 2022.