Combining Tweets and Connections Graph for FakeNews Detection at MediaEval 2022

Published in Multimedia Benchmark Workshop 2022, 2022

The FakeNews Detection task at MediaEval 2022, running for the third time as part of the challenge, focuses on the detection of misinformation tweets and their spreaders. Like in the 2021 task, conspiracy theories related to COVID-19 in nine different categories have to be detected, along with the authors stance towards them. For the 2022 challenge, the size of the dataset has approximately doubled. Furthermore, we also provide a large interaction graph along with vertex features derived from the same Twitter dataset in which misinformation spreaders should be classified. As a final subtask, participants are asked to combine text and graph information to refine their classifications. This paper describes the tasks, including use case and motivation, challenges, the dataset with ground truth, the required participant runs, and the evaluation metrics.

Download paper here


  title={Combining tweets and connections graph for fakenews detection at mediaeval 2022},
  author={Pogorelov, Konstantin and Schroeder, Daniel Thilo and Brenner, Stefan and Maulana, Asep and Langguth, Johannes},
  booktitle={Multimedia Benchmark Workshop},