Understanding and stopping the spread of fake news

All eyes on 2022 success stories, researchers of the Luxembourg Institute of Science and Technology used machine-learning techniques to process large quantities of data with the aim of identifying suspicious trends, signals and behaviour, as well as entities that are behaving badly.

The sharing of misleading information, whether out of context or completely false, and the banding together to coordinate action ("brigading") that accompanies it, alter the quality of the information we receive online, regardless of whether the bad behaviour was intentional.

"CON-NET will push back the frontiers of multilayer network visualisation to allow an end user to better understand the source and impact of online misbehaviour and misinformation," Fintan MC GEE, Senior Research and Technology Associate

Through the CON-NET project, LIST uses machine-learning techniques to process large quantities of data with the aim of identifying suspicious trends, signals and behaviour, as well as entities that are behaving badly. This approach is complemented by a visual analysis approach, including the human factor in the loop, in order to provide a context and understand the spread of misinformation online. In this way, we tackle the complexity of social media networks online.

The CON-NET project is funded within the framework of the CHIST-ERA programme and brings together a consortium of partners from all over Europe.

Discover more success stories in the digital version of the LIST 2022 annual report.