A Dual-Framework Approach for Fake News Detection Using Transformer-Based Embeddings and Explainable AI

DOI

10.2991/978-94-6239-654-8_45How to use a DOI?

Keywords

Misinformation Detection; Natural Language Processing; Deep Learning; Explainable AI; BERT; ss Content Verification

Abstract

The spread of incorrect information across digital media is a significant barrier to user confidence in the information and trustworthiness of digital content. In order to assist content users in spotting false information in online content, this article proposes a comprehensive dual framework that relies on social networks’ community relational characteristics (graph-based propagation), as well as on where the content is shared (contextual embedding). The proposed architecture consists of multiple elements that include fine-tuning from the BERT model, transformer neural network structure to produce semantic representations of text-based content; weighted probability aggregators (WMA) for how information propagates within a community in the social network; attention visualizations; local interpretable model agnostic explanations (LIME); multi-layered approaches for evaluating the accuracy of information; and state-of-the-art pre-processing techniques like tokenization, lemmatization, and extracting contextual features, which provide the ability to build flexible annotation formats (binary, multi-class, and severity) for incorrectly identified media. We present experimental validation showing our proposed system outperforms other methods across multiple benchmark datasets regarding precision, accuracy, recall, and F1-score, quadrupling the baseline systems’ performance. Lastly, this deployment-ready approach supports RESTful APIs for on-demand content verification.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - M. Roshan
AU  - K. P. Monish
AU  - J. A. Adlin Layola
AU  - Kiruba Wesley
PY  - 2026
DA  - 2026/04/24
TI  - A Dual-Framework Approach for Fake News Detection Using Transformer-Based Embeddings and Explainable AI
BT  - Proceedings of the Global Conference on Sustainable Energy Systems, Smart Electronics and Intelligent Computing (GCSESEIC 2025)
PB  - Atlantis Press
SP  - 561
EP  - 573
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-654-8_45
DO  - 10.2991/978-94-6239-654-8_45
ID  - Roshan2026
ER  -