Combating disinformation online and its destabilizing and polarizing social effects


The DeFaktS project takes a comprehensive approach to researching and combating disinformation. For this purpose, an artificial intelligence (AI) is trained based on extracted messages from suspicious social media and messenger groups, which can recognize factors and stylistic devices that are characteristic of disinformation. The AI ​​trained in this way forms a component for an XAI (Explainable Artificial Intelligence). With the help of this, an app should transparently inform and warn users of online offerings about the possible occurrence of disinformation so that the users' media skills are also trained.


With the help of an XAI (Explainable Artificial Intelligence), disinformation on the Internet is to be recognized and users can be made aware of it. For this usage, an AI is trained to recognize the stylistic devices of disinformation. Online platforms should be able to be moderated automatically using a programming interface for third parties.



Liquid Democracy:

University Marburg:

Murmuras GmbH

Murmuras collects consumption and advertising-related data, using its own in-app technology. Murmuras also researches the consumption of technical media. Due to the technical complexity and required interoperability, development is carried out by Murmura's senior developers in close collaboration with the University of Marburg.


Liquid Democracy e.V. is an association that works on projects involving digital participation and democratic participation in non-profit work. Liquid Democracy tests the use of the XAI components to be developed in DeFaktS to detect fake news in a practice-oriented manner and in a real environment in the context of online political debates.
Due to its expertise in the development of data extraction methods and algorithms, the Philipps University of Marburg is an important partner in the DeFaktS project and makes the data sets available via APIs developed by Murmuras, which forms the data basis for the machine learning process.