Fake news articles and pseudoscientific views spread on the internet with little constraint and approximately ten times faster than real news. This puts the roughly 5 billion people who use the internet in enormous danger, especially the 4.7 billion people who also use social media. It also became clear how much damage they could cause when it came to the COVID-19 pandemic. According to the report released on the 27th of April 2021 by the World Health Organization (WHO), approximately 6000 people across the world ended up in hospital and 800 people lost their lives in just the first 3 months of 2020 due to fake news; in the United States, as a result of large quantities of disinformation, the number of deaths and infections has in fact increased. Pandemic-related fake news also contributed to economic damages such as the attacks against 5G towers in Western Europe. False information around the Russia-Ukraine war is also being pushed at an unprecedented rate across the world, the consequences of which are yet to be seen. Since fake news and pseudoscientific views are widespread in a number of different fields, their identification is, evidently, paramount from both a societal and economical perspective. Recognizing the extent of the damages caused by and the threat presented by the dissemination, intentional or otherwise, of disinformation, the Council of the European Union, starting in 2015, has been fighting against the circulation of disinformation and against the resulting consequences by creating and supporting the operations of a variety of action plans, alarm systems, observatories, fact-checking services, research projects etc. (SOLTÉSZ 2023). In order to coordinate such activities of European member states, the European Digital Media Observatory (EDMO) was formed in 2020 by the European Commission. The Hungarian Digital Media Observatory (Magyar Digitális Média Obszervatórium, HDMO), in an expanded form, with the participation of six autonomous organizations, joined EDMO in 2023. Out of the six, Lakmusz and AFP perform fact-checking operations.

According to their methodology, the fake news detection tools known thus far can be sorted into three categories:

1. Lists of sites that share fake news such as lists of intentionally deceiving sites created by HVG and Urban Legends.
2. Fact-Checking such as Oigetit Fake News Filter; PolitiFact; Lakmusz.
3. NLP-based text categorization methods (NLP = Natural Language Processing).

The former two require human decision-making which makes them immensely slow and costly, as a result, they are only used to evaluate a limited number of specific pieces of text and websites. The third method, however, has the capability to analyze even longer than usual pieces of text in real-time, using the frequency of words (e.g. TF-IDF (term frequency-inverse document frequency)). At the same time, if one only takes into account the vocabulary of a given piece of writing, that can easily lead to misidentification, as an example: finding the words covid, vaccine and chip together in an article might not actually mean that it is fake news, it might simply be a rebuttal to unscientific views. 

Thus, the question presents itself: how can fake news and pseudoscientific articles be reliably identified, and do they possess attributes that can be used to identify them?

Our research group's stated goal is to answer these questions, placing our research into a broader, scientific, linguistics-based scientific perspective encompassing both basic and applied research, as well as innovational perspective. Our starting hypothesis is that fake news and pseudoscientific articles contain linguistic traits and strategies based on which or on their combination it is possible to label them as probable fake news or pseudoscience.

Our research has three main goals:

• Establishing a basis for the scientific research of fake news and pseudoscientific views.
• Unveiling the linguistic properties and strategies of fake news and pseudoscientific views in the field of health and medicine.
• Developing tools capable of automatically identifying fake news and pseudoscientific views within the field of health and medicine, in the form of a smartphone application for Android and iOS, as well as a browser extension.