Organisation Profile
The Natural Language Processing (NLP) Group at the University of Sheffield is one of the largest and most successful research groups in text mining and language processing in the EU. The group is based in the Department of Computer Science, which also includes world-class teams in the areas of speech, knowledge and information processing, biotechnology, and machine learning for medical informatics.
The NLP group has world-leading research records in the fields of social media analysis (in particular rumours and misinformation), machine learning methods for NLP, intelligent conversational agents and dialogue systems, NLP infrastructures (GATE), cloud-based text-analytics-as-a-service, information extraction, question answering, and terminology extraction.
The WeVerify project will build on USFD’s strong misinformation analysis expertise, initially gathered through leading the PHEME project, and now continued in the SoBigData and COMRADES projects. The PHEME project, alongside REVEAL, was among the first world-wide to study computational methods for detecting and tracking mis- and disinformation in social media. We coined the term pheme to describe memes which are enhanced with truthful information.
Other relevant previous EC and national projects of relevance to WeVerify include COMPANIONS (intelligent digital companions), CLARIN ERIC (language resources infrastructure), AnnoMarket (cloud-based NLP services), GateCloud (UK JISC project), EnviLOD (UK JISC project on mining scientific collections with the British Library).
USFD also has extensive experience in innovation and knowledge transfer activities, in particular collaborations and consulting for companies (both SMEs and large corporations), government bodies, and other organisations.
Role in the Project
USFD will lead WP6 which will develop the WeVerify citizen-oriented digital companion. In WP4 the project team will develop human-in-the-loop tools, based on state-of-the-art machine learning-based algorithms from the PHEME project (including rumour detection, stance analysis, and veracity classification). USFD will also engage actively with the definition and implementation of the open-source WeVerify platform (WP3), as well as contribute to the use case pilots, dissemination, and community building.