Skip to main content
EXTENDED SEARCH
Log in
Top

Investigating Fake News Detection Using BERT/RoBERTa LLMs

  • 2025
  • OriginalPaper
  • Chapter
Published in:
Authors
Amr Abu Alhaj
Omar Safwat
Youssef Ghoneim
Imran Zualkernan
Ali Reza Sajun
Published in
Proceedings of Tenth International Congress on Information and Communication Technology
Publisher
Springer Nature Singapore

Activate our intelligent search to find suitable subject content or patents.

search-config
loading …

Abstract

This chapter delves into the critical task of detecting fake news, particularly in political discourse, using advanced transformer models like BERT and RoBERTa. The study introduces a versatile architecture that combines these models with BiLSTM and CNN layers, aiming to capture both local and global contexts of political statements. A key innovation is the incorporation of a credit score feature, which quantifies a speaker's historical truthfulness, adding a temporal context to the model's decision-making process. The research compares the performance of different model configurations, highlighting the superior accuracy achieved by the BERT BiLSTM + CNN architecture. The study also explores the use of GeLU activation functions, which outperform ReLU in capturing complex data relationships. Results show that BERT consistently outperforms RoBERTa in this specific task, achieving state-of-the-art performance on the LIAR dataset. The chapter concludes with a discussion on the potential for further optimization and the importance of safeguarding public discourse against misinformation.
AI Generated

This summary of the content was generated with the help of AI.

Please log in to get access to your license.

Log in Register for free

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Business + Economics & Engineering + Technology"

Online-Abonnement

Springer Professional "Business + Economics & Engineering + Technology" gives you access to:

  • more than 102.000 books
  • more than 537 journals

from the following subject areas:

  • Automotive
  • Construction + Real Estate
  • Business IT + Informatics
  • Electrical Engineering + Electronics
  • Energy + Sustainability
  • Finance + Banking
  • Management + Leadership
  • Marketing + Sales
  • Mechanical Engineering + Materials
  • Insurance + Risk


Secure your knowledge advantage now!

inform now

Springer Professional "Engineering + Technology"

Online-Abonnement

Springer Professional "Engineering + Technology" gives you access to:

  • more than 67.000 books
  • more than 390 journals

from the following specialised fileds:

  • Automotive
  • Business IT + Informatics
  • Construction + Real Estate
  • Electrical Engineering + Electronics
  • Energy + Sustainability
  • Mechanical Engineering + Materials





 

Secure your knowledge advantage now!

inform now

Springer Professional "Business + Economics"

Online-Abonnement

Springer Professional "Business + Economics" gives you access to:

  • more than 67.000 books
  • more than 340 journals

from the following specialised fileds:

  • Construction + Real Estate
  • Business IT + Informatics
  • Finance + Banking
  • Management + Leadership
  • Marketing + Sales
  • Insurance + Risk



Secure your knowledge advantage now!

inform now
previous chapter Navigating Market Volatility in Agriculture Through Digital Analytics: Lessons from Indonesia’s Coffee Sector
next chapter Multi-level Hybrid Ensemble with Attention-Based Meta-Learner for Diabetic Retinopathy Prediction
Title
Investigating Fake News Detection Using BERT/RoBERTa LLMs
Authors
Amr Abu Alhaj
Omar Safwat
Youssef Ghoneim
Imran Zualkernan
Ali Reza Sajun
Copyright Year
2025
Publisher
Springer Nature Singapore
Book
Proceedings of Tenth International Congress on Information and Communication Technology

Print ISBN: 978-981-9669-28-8

Electronic ISBN: 978-981-9669-29-5

Copyright Year: 2025

DOI
https://doi.org/10.1007/978-981-96-6929-5_19
This content is only visible if you are logged in and have the appropriate permissions.