Scholarlink

<p>With the rapid growth of Internet technology and the increasing reliance on digital platforms for information, the spread of fake news has become a serious global concern. Fake news undermines public trust, distorts informed decision-making, and threatens the stability of key sectors, including healthcare, politics, and the economy. In the Arabic context, challenges such as linguistic complexity, dialectal variation, and the limited availability of annotated datasets make detection particularly difficult, highlighting the need for focused research. We present the first comprehensive survey covering all relevant papers published between 2019 and 2025, examining available datasets with respect to their collection, annotation methodologies, topical coverage, and limitations. Additionally, we categorize detection approaches according to classification paradigms including machine learning, deep learning, transformer-based models, and other strategies. We analyze discriminative features used to differentiate fake from real news, organizing them into content-based and context-based groups. Finally, we identify key research gaps and outline future directions, aiming to provide a solid foundation for developing more robust and adaptable Arabic fake news detection systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Fake news detection in Arabic: a survey of linguistic challenges, resources, and classification approaches

  • 期刊论文

  • 发表日期：

    2026年5月28日

查看全文
  • Meriem Sellami,
  • Nadjet Kamel,
  • Abdelaziz Lakhfif

摘要

With the rapid growth of Internet technology and the increasing reliance on digital platforms for information, the spread of fake news has become a serious global concern. Fake news undermines public trust, distorts informed decision-making, and threatens the stability of key sectors, including healthcare, politics, and the economy. In the Arabic context, challenges such as linguistic complexity, dialectal variation, and the limited availability of annotated datasets make detection particularly difficult, highlighting the need for focused research. We present the first comprehensive survey covering all relevant papers published between 2019 and 2025, examining available datasets with respect to their collection, annotation methodologies, topical coverage, and limitations. Additionally, we categorize detection approaches according to classification paradigms including machine learning, deep learning, transformer-based models, and other strategies. We analyze discriminative features used to differentiate fake from real news, organizing them into content-based and context-based groups. Finally, we identify key research gaps and outline future directions, aiming to provide a solid foundation for developing more robust and adaptable Arabic fake news detection systems.