EUVIMEDDécouvrir EUVIMED
Open access · CC BY 4.0

Fake News Detection Through LLM-Driven Text Augmentation Across Media and Languages

Abdul Sittar, Mateja Smiljanic, Alenka Guček, Marko Grobelnik

Machine Learning and Knowledge Extraction · 2026-04-15 · 8 (4) · 103

DOI: 10.3390/make8040103 ↗

At a glance

Journal article in Machine Learning and Knowledge Extraction, published online on 2026-04-15. Written by Abdul Sittar and 3 co-authors. Cited 1 times so far (Crossref); 50 references. The full text is freely available (open access, CC BY 4.0). Funded by: European Union under the Horizon Europe framework.

RésuméRéférences bibliographiquesCiteRelated studies

Abstract (original language)

The proliferation of fake news across social media, headlines, and news articles poses major challenges for automated detection, particularly in multilingual and cross-media settings affected by data imbalance. We propose a fake news detection framework based on LLM-driven, feature-guided text augmentation. The method generates realistic synthetic samples across languages, media types, and text granularities while preserving meaning and stylistic coherence. Experiments with classical and transformer-based models (Random Forest, Logistic Regression, BERT, XLM-R) across social media, headlines, and multilingual news datasets show consistent improvements in performance. For inherently balanced datasets (e.g., social media), synthetic augmentation yields negligible but stable performance changes. Across imbalanced scenarios, synthetic augmentation substantially improves minority-class recall and F1-score (e.g., fake news recall from 0.57 to 0.86), while preserving majority-class performance, leading to more balanced and reliable classifiers, whereas oversampling significantly degrades results due to overfitting on duplicated language patterns. Overall, a hybrid semantic- and style-based model proves to be the most robust strategy, outperforming oversampling and matching or exceeding baseline performance across datasets.

Funding

  • European Union under the Horizon Europe framework

Cite

APA

Sittar, A., Smiljanic, M., Guček, A., & Grobelnik, M. (2026). Fake News Detection Through LLM-Driven Text Augmentation Across Media and Languages. Machine Learning and Knowledge Extraction, 8(4), 103. https://doi.org/10.3390/make8040103

Vancouver

Sittar A, Smiljanic M, Guček A, Grobelnik M. Fake News Detection Through LLM-Driven Text Augmentation Across Media and Languages. Machine Learning and Knowledge Extraction. 2026;8(4):103. doi:10.3390/make8040103

BibTeX
@article{Sittar2026,
  author = {Sittar, Abdul and Smiljanic, Mateja and Guček, Alenka and Grobelnik, Marko},
  title = {Fake News Detection Through LLM-Driven Text Augmentation Across Media and Languages},
  journal = {Machine Learning and Knowledge Extraction},
  year = {2026},
  volume = {8},
  number = {4},
  pages = {103},
  doi = {10.3390/make8040103}
}
Start for free →