Duration: 2024-2025
Conductor: Prof. Dr.-Ing. Christian Grimme (University of Münster), Prof. Dr. Thorsten Quandt (University of Münster)
Institutions: Institute of Information Systems (University of Münster), Department of Communication (University of Münster)
Funding: Ideenlabor, University of Münster
Keywords: Disinformation, Disinformation campaigns, Fake News, Public opinions, Computational Methods, Image analysis
Recent research in the field of online communication has increasingly focused on the negative effects of social media usage in recent years. Particularly, the spread of disinformation during crises such as the COVID-19 pandemic and the Ukraine conflict has been of interest, as the risks to liberal democracies have become abundantly clear.
A significant challenge for research lies in the analysis of images and videos that are often included in communication data. Current techniques for image analysis require extensive image information and are often unable to process the vast amounts of data in real-time.
The presented project pursues an innovative idea to address these challenges. Instead of analyzing the complete image content, image analysis should be optimized using human attention mechanisms. The use of this technique would significantly improve the efficiency of image processing in data streams. The goal of the project is to develop a prototype that can be used in future research projects on disinformation and security research.