Naslov
Deep Learning with Self-Attention Mechanism for Fake News Detection

Autori
Cvitanović, Ivana ; Bagić Babac, Marina

Vrsta, podvrsta i kategorija rada
Poglavlja u knjigama, znanstveni

Knjiga
Combating Fake News with Computational Intelligence Techniques

Urednik/ci
Lahby M. ; Pathan AS.K. ; Maleh Y. ; Yafooz W.M.S.

Izdavač
Springer

Godina
2022

Raspon stranica
205-229

ISBN
978-3-030-90086-1

Ključne riječi
Natural language processing, Fake news, Self-attention, Deep learning, Transfer learning

Sažetak
Nowadays, fake news is one of major concerns in our society, that is a form of news consisting of deliberate disinformation or hoaxes spread via traditional news media or online social media. Thus, this study aims to explore state-of-the-art methods for detecting fake news in order to design and implement classification models. Four different classification models based on deep learning with self-attention mechanism were trained and evaluated using current datasets that are available for this purpose. Three models explored traditional supervised learning, while the fourth model explored transfer learning by fine-tuning the pre-trained language model for the same task. All four models yield comparable results with the fourth model achieving the best classification accuracy.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA