Centre for Language and Speech Technology (CLST) is part of international project to combat disinformation

Date of news: 25 January 2023

Humans and machines will join forces to curb one of today’s biggest threats to democracy: fake news, hate speech and misinformation. They will face a new obstacle in the European consortium 'HYBRIDS': an international initiative funded with over 2.9 million euros. From Radboud University, researchers of the Centre for Language and Speech technology (CLST) are involved in the project.

Fake news

The spread of rumours and fake news, hate speech against vulnerable minorities and harassment through social media have become one of the main threats to modern democracies. Parliaments and governments of different nations are becoming aware of the need for a global strategy against disinformation, assuming that the stability of democratic societies also depends on citizens having access to verifiable information, thus being free to construct their own opinions based on objective data whose veracity can be demonstrated. Although there have been significant advances in the automatic detection of misinformation in recent years, it remains an under-studied field, as it is an extremely complex task that requires deep semantic knowledge and different inference and reasoning mechanisms based on natural language.

New tools

The European Union and UK Research and Innovation (UKRI) are supporting a new project to stop the spread of fake news and hate speech through social media. The project, called HYBRIDS, will use over 2.9 million euros from the EU's Horizon Europe research and innovation program and UKRI's Horizon Europe funding guarantee to design and develop new tools using natural language processing and artificial intelligence.

The project will focus on analyzing public discourse about important global issues such as the climate crisis, immigration, and European skepticism, and will use deep learning techniques to create a "hybrid intelligence" that combines the abilities of computer algorithms with human knowledge. The project will also help train people in the field of hybrid methodologies to help improve democratic practices in Europe.

Team: academia and industry

From its very initial conception, HYBRIDS has assumed the need for a multidisciplinary research team, capable of transferring the knowledge acquired in the human and social sciences to the technological tools of Natural Language Processing (NLP) and Deep Learning (DL) algorithms.

HYBRIDS is the result of an international consortium led from Santiago de Compostela (Spain) by the Research Centre on Intelligent Technologies (CiTIUS), and it involves a total of 14 members (8 beneficiaries - 6 partners) from 7 European countries: France, Germany, Italy, Portugal, Spain, the Netherlands and the United Kingdom.

The project started on 1 January 2023.