About the Project

The project is funded by the BMBF Research Programme on Detecting and Combating Digital Disinformation Campaigns. The core idea of the project is to analyse the narrative structures of disinformation in videos from public media and alternative media shared on social networks.

Narrative structures of disinformation in public media are highly dangerous, as they fuel fundamental distrust among the audience and thus drive them into the arms of alternative, deliberately misleading information channels. Critical questions thus arise about similarities and differences between public media sources and alternative news videos of targeted disinformation, which will be investigated in this project.

In this project, we combine an innovative way of computer-assisted analysis and visualization approaches with digital humanities and multimodal semiotics to develop methods for narrative structure and argument construction analysis.

Three concrete goals of the project are:

• Semi-automatic Identification of Typical Narratives of Disinformation: in alternative news videos as well as public media broadcasts by using state-of-the-art machine learning methods (Computer Vision, Natural Language Processing), Visual Analytics and multimodal video analysis.
• Development of an Interactive Software Tool: it can be used by experts and the general public in a simple way to automatically recognize and understand basic narrative strategies.
• Development of Best Practices: for the creators of public media broadcasts to avoid unintentional narratives of disinformation, and to neutralize existing disinformation campaigns by targeted counter-messages.