Fake News Detection

Marín, Ignacio Palacio; Arroyo, David

doi:10.1007/978-3-030-57805-3_22

Cited by 5 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has gained large interest in the last five years due to the increasing circulation of misinformation in social media [Pérez-Rosas et al 2018, Marín and Arroyo 2019, Oshikawa et al 2020. Despite its simplicity and conceptual limitations, approaches for fake news detection based on classification considering simple linguistic features, such as lexical and stylistic characteristics of deceptive language use, have achieved considerable accuracy in the available datasets [Oshikawa et al 2020, Al-Rakhami andAl-Amri 2020].…”

Section: Covid-related Fake News Detection Through Text Classificationmentioning

confidence: 99%

Fake News Detection in Tweets: Challenges and Adaptations imposed by the COVID-19

Santana

Claro

Souza

2022

iSys

View full text Add to dashboard Cite

Misinformation has plagued citizens’ lives, especially on social networks. During the COVID-19 pandemic, the proliferation of competing narratives and dissemination of false or inaccurate news about the pandemic has reached such a state that led the World Health Organization to classify it as an infodemic. However, few resources are available to combat misinformation in this new and evolving domain, especially considering how social networks allow the rapid spreading of false narratives. In this case, the lack of resources, such as methods, tools, and reliable information on the virus, hinders our ability to combat this misinformation. In this work, we investigate the application of Text Analysis methods to help health-related scientific communicators produce educational material to combat misinformation. This study was conducted in association with the Scientific Communication sector of FIOCRUZ, a health research institution in Brazil, aiming to monitor COVID-19-related fake news and produce educational material to combat misinformation in a weekly manner due to the ephemeral nature of COVID-19 misinformation in social media. As the main findings of this work, we provide (1) a pipeline for automatically collecting and analyzing news and social media posts regarding COVID-19 in orderto provide science communicators with a weekly contextualized view of topics related to COVID-19 in social media; (2) we analyzed the effect of different resources and methods in the analytical tools employed in this work for detecting health-related misinformation in the Portuguese language, and finally, (3) we provided to journalists and science communicators in FIOCRUZ computational tools to automatically monitor COVID-related misinformation in social media, focusing on Twitter, aiming to contribute to definition of the weekly science communication agenda of the institution. Indeed, we indicate the type of resources to combat misinformation in the pandemic, and our approach can handle the detection of misinformation on Twitter social networks within the COVID-19 domain.

show abstract

Section: Covid-related Fake News Detection Through Text Classificationmentioning

confidence: 99%