Affiliation:
1. Politehnica University of Bucharest
Abstract
Fighting fake news is a difficult and challenging task. With an increasing impact on the social and political environment, fake news exert an unprecedently dramatic influence on people’s lives. In response to this phenomenon, initiatives addressing automated fake news detection have gained popularity, generating widespread research interest. However, most approaches targeting English and low-resource languages experience problems when devising such solutions. This study focuses on the progress of such investigations, while highlighting existing solutions, challenges, and observations shared by various research groups. In addition, given the limited amount of automated analyses performed on Romanian fake news, we inspect the applicability of the available approaches in the Romanian context, while identifying future research paths.
Publisher
ASTRA National Museum Complex
Subject
Literature and Literary Theory,History,Cultural Studies
Reference26 articles.
1. Allcott, Hunt, Matthew Gentzkow, and Chuan Yu. “Trends in the Diffusion of Misinformation on Social Media.” Research and Politics 6, no. 2 (2019): 1-8.
2. Bhatt, Gaurav, Aman Sharma, Shivam Sharma, Ankush Nagpal, Balasubramanian Raman, and Ankush Mittal. “Combining Neural, Statistical and External Features for Fake News Stance Identification.” In WWW ’18: Companion Proceedings of the The Web Conference, 1353–1357. International World Wide Web Conferences Steering Committee Republic and Canton of Geneva, Switzerland, 2018.
3. CNN. Retrieved November 2, 2020, from https://edition.cnn.com/
4. Crammer, Koby and Yoram Singer. “On the Algorithmic Implementation of Multiclass Kernel-based Vector Machines.” Journal of Machine Learning Research 2 (2001): 265–292.
5. Devlin, Jacob, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. “Bert: Pre-training of Deep Bidirectional Transformers for Language Understanding.” In Proceedings of NAACL-HLT 2019, 4171–4186. Minneapolis, Minnesota: ACL, 2019.
Cited by
9 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献