We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A Novel Transformer-Based Deep Learning Pipeline for Multilingual Fake News Detection.
- Authors
Agras, Kagan; Atay, Begum
- Abstract
The emergence of extensive amounts of fake news on the internet made real-time fake news detection (FND) through computational tools necessary. However, the non-availability of annotated data and language-specific processing tools significantly undermines the application of FND in low-resource languages. This study introduces a transformer-based pipeline for detecting fake news and expanding annotated datasets in low-resource languages. A multilingual dataset containing 16 languages was created using machine translation. Multilingual Bidirectional Encoder Representations from Transformers (mBERT) was fine-tuned on the combined dataset and demonstrated an accuracy rate of 97% in FND. Our model could learn languageindependent features through cross-lingual language understanding (XLU) and perform similarly to monolingual models. Experiments performed on zero-shot settings indicate that our model can accurately perform FND in languages not included in the training set. To increase accessibility and practicality, the model is deployed on a Chrome web extension developed with JavaScript. The data input to the extension is collected in a dataset. Impractical data are removed using a filtering model. Overall, this study offers a novel self-expanding pipeline for FND in all languages and the expansion of annotated data in real-time, which is a significant step towards overcoming public misinformation on a global scale.
- Subjects
DEEP learning; FAKE news; JAVASCRIPT programming language; NATURAL language processing; MACHINE learning
- Publication
International Journal of High School Research, 2024, Vol 6, Issue 4, p73
- ISSN
2642-1046
- Publication type
Article
- DOI
10.36838/v6i4.12