Warsaw University of Technology, in collaboration with partners from Japan and Spain, is developing tools to effectively fight fake news in the virtual world.

Real events presented in a wrong context, randomly forwarded and unverified information, manipulated facts, or deliberate disinformation actions as a component of hybrid warfare - fake news has become a real concern. Social media platforms are particularly vulnerable to misinformation.

To counteract this, researchers from Warsaw University of Technology, Okayama University (Japan), and Fundació per a la Universitat Oberta de Catalunya (Spain) are implementing the DISSIMILAR project.

– ‘We are currently focusing on social issues that will help us better understand the mechanisms of creating and spreading false information. – says Anna Visvizi, PhD, SGH Professor, a member of the DISSIMILAR team. – In the next phase of the project, we will present methods to counteract the spread of fake news based on the most recent techniques using artificial intelligence’ – she adds.

Our researchers are working on IT tools that, when combined with data from international social research, will serve as the foundation for new software to combat disinformation in social media.

– ‘We are currently developing a virtual platform for testing watermarking, machine learning, and computer forensics mechanisms, all of which will be used to combat fake news. On this basis, we will build an efficient tool that will automatically separate fake information and prevent its spread’ – emphasizes Wojciech Mazurczyk, PhD, DSc, WUT Professor from the Faculty of Electronics and Information Technology, who is directing the Polish team in the DISSIMILAR project.

The project's results will be useful in other areas as well. The employment of digital marking techniques, that is, embedding so-called watermarks in protected content, which are invisible and inaudible, and their removal is immediately noticeable, will provide creators with an effective tool for copyright protection.

More information about the DISSIMILAR (Detection of fake newS on SocIal MedIa pLAtfoRms) project can be found on its dedicated website.