Login Search
FMDB Transactions on Sustainable Computer Letters Submit Paper
    Crossref logo

    Archiving

    Portico

    Fernando Martins De Bulhão is a member of

    Crossref logo

A Smart Approach for Precise Fake News Detection Using Bi-Directional LSTM and Self-Attention Mechanism

Authors:
O. Jeba Singh, R. Remya, G. Dhivyasri, S. Manikandan, S. Rubin Bose

Addresses:
Center for Academic Research, Alliance University, Bengaluru, Karnataka, India. Department of Electronics and Communication Engineering, Sri Krishna College of Engineering and Technology, Coimbatore, Tamil Nadu, India. Department of Computer Science and Engineering, Sai Vidya Institute of Technology, Bengaluru, Karnataka, India. Department of Electronics and Communication Engineering, Presidency University, Bengaluru, Karnataka, India. School of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India.

Abstract:

The legitimacy of information and the trust of the public are both in jeopardy because of the swift rise in the amount of false information that is disseminated through social media and other computerised platforms. Statistical characteristics are the primary foundation for most conventional techniques that are used to detect false information in the news. However, these characteristics often fail to account for the intricate variations in semantics and context found in news stories. This paper introduces an enhanced fake news detection model that blends Bi-directional Long Short-Term Memory (Bi-LSTM) networks with a self-attention mechanism to address this restriction. Bi-LSTM networks analyse text in both forward and backward directions, enabling them to capture long-range contextual correlations within the content. The self-attention mechanism is an additional feature that is incorporated into the model to improve its performance. It does this by dynamically assigning distinct words and phrases in the text to different levels of priority. The Kaggle fake news dataset, also known as the ISOT public dataset, is used to evaluate the model's performance. 

Keywords: Fake News Detection; Bi-LSTM Networks; Self-Attention; Deep Learning; Data Ecosystem; Low Detection; Nuanced Cases; Long Short-Term Memory (LSTM); Manipulated News.

Received on: 25/12/2024, Revised on: 02/03/2025, Accepted on: 14/04/2025, Published on: 22/11/2025

DOI: 10.69888/FTSCL.2025.000484

FMDB Transactions on Sustainable Computer Letters, 2025 Vol. 3 No. 4, Pages: 173-182

  • Views : 26
  • Downloads : 11
Download PDF