| Revue: | Computación y sistemas |
| Base de datos: | |
| Número de sistema: | 000560624 |
| ISSN: | 1405-5546 |
| Autores: | Shushkevich, Elena1 Alexandrov, Mikhail2 Cardiff, John1 |
| Instituciones: | 1Technological University Dublin, Dublin. Irlanda 2Russian Presidential Academy of National Economy and Public Administration, Moscow. Rusia 3Universitat Autónoma de Barcelona, Barcelona, Cataluña. España |
| Año: | 2021 |
| Periodo: | Oct-Dic |
| Volumen: | 25 |
| Número: | 4 |
| Paginación: | 783-792 |
| País: | México |
| Idioma: | Inglés |
| Resumen en inglés | The increase of fake news in social media, especially about Covid-19, poses a real threat to the mental and physical health of people. It is an important task to detect such news and to stop it spreading. In this article, we describe the main approaches for fake news about Covid-19 detection, including Classical Machine Learning models, models based on Neural Networks and models, which were created based on the other approaches and preprocessing steps. We analyze the results of the challenge “Constraint@AAAI2021 -COVID19 Fake News Detection”, the main goal of which was the binary classification of news collected from social media for fake and real news. We analyze the best approaches, which were proposed by researchers during the challenge. In addition, we describe datasets of fake news related to Covid-19, which could be useful for the detection and classification of such news. |
| Keyword: | Fake news, Covid-19, Classical machine learning models, Neural networks, Text transformers |
| Texte intégral: | Texto completo (Ver HTML) Texto completo (Ver PDF) |
