EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hybrid Feature and Optimized Deep Learning Model Fusion for Detecting Hateful Arabic Content

Karim Gasmi, Ibtihel Ben Ltaifa, Alameen Eltoum Abdalrahman, Omer Hamid, Mohamed Othman Altaieb, Shahzad Ali, Lassaad Ben Ammar, Manel Mrabet

IEEE Access · 2025 · 13 · 131411-131431

DOI: 10.1109/access.2025.3591673 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2025. Verfasst von Karim Gasmi und 7 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 103 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Graduate Studies and Scientific Research at Jouf University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detecting hate speech in Arabic social media content is critical for ensuring safe, inclusive, and respectful online communication. However, this task remains challenging due to Arabic’s morphological richness, dialectal variations such as Levantine, and the scarcity of high-quality annotated data. This study proposes a comprehensive and language-aware approach to Arabic hate speech detection that integrates advanced preprocessing, targeted data augmentation, hybrid feature extraction, and deep ensemble learning. Our experiments are conducted on a Levantine Arabic tweet dataset labeled hateful or non-hateful. To address lexical variability and noise common in user-generated content, we apply a dedicated preprocessing pipeline that includes normalization, diacritic removal, and emoji filtering. To further enhance generalization and mitigate data imbalance, we employ two augmentation strategies: synonym replacement using a curated Arabic lexicon and semantic-preserving back-translation through English. We investigate lexical and contextual approaches for feature extraction, including TF-IDF vectors, contextualized AraBERT embeddings, and a hybrid combination of both. These features are input into multiple deep learning classifiers, including CNN-BiGRU, BiLSTM, and DNN architectures. To maximize predictive performance, we develop an ensemble framework that integrates these models. The final prediction is obtained through a weighted fusion of individual model outputs, where the optimal weights are selected using the Grey Wolf Optimizer (GWO), aiming to maximize classification accuracy. Experimental results demonstrate that our proposed hybrid and ensemble-based architecture achieves superior performance, with an accuracy of 83.33% and a ROC-AUC score of 89.5%, outperforming individual models and conventional baselines. These findings highlight the effectiveness of hybrid feature representations and nature-inspired optimization in enhancing Arabic hate speech detection. Our approach offers a scalable, linguistically informed solution for robust content moderation in Arabic digital spaces.

Förderung

  • Deanship of Graduate Studies and Scientific Research at Jouf University

Zitieren

APA

Gasmi, K., Ltaifa, I. B., Abdalrahman, A. E., Hamid, O., Altaieb, M. O., Ali, S., Ammar, L. B., & Mrabet, M. (2025). Hybrid Feature and Optimized Deep Learning Model Fusion for Detecting Hateful Arabic Content. IEEE Access, 13, 131411-131431. https://doi.org/10.1109/access.2025.3591673

Vancouver

Gasmi K, Ltaifa IB, Abdalrahman AE, Hamid O, Altaieb MO, Ali S, et al. Hybrid Feature and Optimized Deep Learning Model Fusion for Detecting Hateful Arabic Content. IEEE Access. 2025;13:131411-131431. doi:10.1109/access.2025.3591673

BibTeX
@article{Gasmi2025,
  author = {Gasmi, Karim and Ltaifa, Ibtihel Ben and Abdalrahman, Alameen Eltoum and Hamid, Omer and Altaieb, Mohamed Othman and Ali, Shahzad and Ammar, Lassaad Ben and Mrabet, Manel},
  title = {Hybrid Feature and Optimized Deep Learning Model Fusion for Detecting Hateful Arabic Content},
  journal = {IEEE Access},
  year = {2025},
  volume = {13},
  pages = {131411--131431},
  doi = {10.1109/access.2025.3591673}
}
Kostenlos starten →