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

    Archiving

    Portico

    Fernando Martins De Bulhão is a member of

    Crossref logo

Advanced Fake News Detection Using BEOSA-Based and an Attentive Convolutional Transformer

Authors:
Rakesh Chandrashekar, Jayasheel Kumar, M. Arunadevi Thirumalraj, S. Gopikha, Prasanna R. Christodoss

Addresses:
Department of Mechanical and Engineering, New Horizon College of Engineering, Bengaluru, Karnataka, India. Department of Computer Science and Engineering, Karunya Institute of Technology and Science, Coimbatore, Tamil Nadu, India. Department of Computer Science and Business Management, Saranathan College of Engineering, Tiruchirappalli, Tamil Nadu, India. Department of Information Technology, St. Joseph's College of Engineering, Chennai, Tamil Nadu, India. Department of Computing, Mathematics and Physics, Messiah University, Mechanicsburg, Pennsylvania, United States of America.

Abstract:

The rapid dissemination of false information, known as "fake news," has been made possible by the advent of social media platforms. False information not only deceives individuals but also fools the community. The spread of false information and the erosion of trust in information sources have harmed individuals and society, driven by the prevalence of poor-quality content on social media platforms. To improve classification accuracy and identify false information, this work utilises a state-of-the-art algorithm and a strict pipeline for preprocessing and feature extraction. In the preprocessing stage, the first step is to eliminate HTML tags, lowercase words, and stop words. The process of feature extraction using TF-IDF and word embeddings can capture more nuanced patterns in language. An innovative Attentive Convolutional Transformer (ACT) model that combines Transformer and CNN architectures is used to detect false information during classification. The Binary Ebola Optimisation Search Algorithm (BEOSA) is used for ACT hyperparameter tuning. Model discrimination and generalisation are both enhanced by BEOSA. 

Keywords: Convolutional Neural Network; Fake News; Social Media; Term Frequency-Inverse Document Frequency; Bogus News; HTML Tags; CNN Architectures; Quick Transmission.

Received on: 25/01/2025, Revised on: 04/04/2025, Accepted on: 16/05/2025, Published on: 22/11/2025

DOI: 10.69888/FTSCL.2025.000487

FMDB Transactions on Sustainable Computer Letters, 2025 Vol. 3 No. 4, Pages: 213-226

  • Views : 13
  • Downloads : 4
Download PDF