Introduction

E-communication in general is increasingly playing crucial roles in everyone's life. Because of that, the analysis of textual information coming from electronic sources has been a popular research topic among the computational linguistics community. Academic competitions or shared tasks that seek to advance the state of the art in a particular research topic (see e.g., the series of events organized by IberLEF, PAN, and SemEval) help to increase the research in such topics providing common evaluation criteria resulting in state of the art methodologies for many NLP related problems. Despite such progress, there are still open issues that deserve further research in order to be solved or at least to better understand them

Fake news provides information that aims to manipulate people for different purposes: terrorism, political elections, advertisement, satire, among others. In social networks, misinformation extends in seconds among thousands of people, so it is necessary to develop tools that help control the amount of false information on the Web.

A fake news detection system aims to help users detect and filter out potentially deceptive news. An approach for the prediction of intentionally misleading news is based on the analysis of truthful and fraudulent previously reviewed news, i.e., annotated corpora.

For the second edition of the fake news detection task, a new corpus containing news associated with COVID-19 will be used as a testing corpus, while the corpus used in the 2019 edition of this task will be provided as the training set for the development of participants’ solutions. Our aim is to explore the robustness of methods when trained on generic news and then evaluated in news associated with a very specific theme.

The Fake New Detection task is organized in collaboration with the MexLef initiative and is a part of the IberLef 2021 shared tasks.