Synthetic Disinformation Data, Infrastructure for Collaboration and Analysis Toolbox for Social Media Research (SynDIKAT)

Basic data for this project

Description

The SynDIKAT project (5/2026–4/2029) addresses the challenges of restricted data access and the lack of standardized benchmarks in social media disinformation research. It aims to develop a framework that facilitates secure data sharing among researchers while ensuring compliance with data protection and licensing. A central component is the generation of realistic synthetic social media datasets using Large Language Models (LLMs) and agent-based simulations to provide a robust basis for evaluating detection algorithms. By establishing a national and international "Data Sharing Community," the collaborative project — led by Prof. Christian Grimme and Prof. Thorsten Quandt (University of Münster) in partnership with GESIS, CAIS, and TU Dortmund — seeks to enhance the transparency and reliability of scientific findings in the field. The project is funded by the German Federal Ministry of Research, Technology and Space.