Scholarlink

<p>With the widespread use of mobile devices and social media platforms, disseminating fake news and misinformation has become a significant concern. Existing research in this area is limited to models developed and tested on datasets curated with limited features, often restricted to specific geopolitical incidents, smaller domains, and typically written only in English (with few exceptions in other languages). To address this gap, we propose the BharatAuthenticNet (BAN) model, developed using an ensemble learning-based approach that effectively operates on a multi-feature, multi-domain, and multi-lingual dataset. This work is the first of its kind to provide a detailed experimental analysis of a proposed model in the field of fake news detection. BAN achieves superior accuracy, precision, recall, and F1 score of 0.9341, 0.9619, 0.9889, and 0.9421, respectively, surpassing baseline algorithms and state-of-the-art techniques. This machine learning-based research specifically contributes to the integrity of social media platforms by mitigating the harmful effects of misinformation. </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

A Novel Ensemble Model BharatAuthenticNet for Detecting Fake News: Experimentation and Performance Analysis

  • 期刊论文

  • 发表日期：

    2025年2月5日

查看全文
  • Manish Kumar Singh,
  • Jawed Ahmed,
  • Kamlesh Kumar Raghuvanshi,
  • Mohammad Afshar Alam

摘要

With the widespread use of mobile devices and social media platforms, disseminating fake news and misinformation has become a significant concern. Existing research in this area is limited to models developed and tested on datasets curated with limited features, often restricted to specific geopolitical incidents, smaller domains, and typically written only in English (with few exceptions in other languages). To address this gap, we propose the BharatAuthenticNet (BAN) model, developed using an ensemble learning-based approach that effectively operates on a multi-feature, multi-domain, and multi-lingual dataset. This work is the first of its kind to provide a detailed experimental analysis of a proposed model in the field of fake news detection. BAN achieves superior accuracy, precision, recall, and F1 score of 0.9341, 0.9619, 0.9889, and 0.9421, respectively, surpassing baseline algorithms and state-of-the-art techniques. This machine learning-based research specifically contributes to the integrity of social media platforms by mitigating the harmful effects of misinformation.