Semantics for Online Misinformation Detection, Monitoring, and Prediction

Workshop at ISWC 2020


The SEMIFORM workshop will take place on Monday 2nd November, 2020 on Zoom. The link is available for the registered participants to the password protected ISWC 2020 program page. The schedule given below is in Central European Time (Paris) or GMT+1.

14:00 - 14:10 Welcome by the chair  
14:10 - 15:00 Keynote Talk Carolina Scarton (University of Sheffield) - WeVerify: Wider and Enhanced Verification for You slides
15:00 - 15:20 Daniel Kazenoff, Oshani Seneviratne and Deborah McGuinness. Semantic Graph Analysis to Combat Cryptocurrency Misinformation on the Web slides
15:20 - 15:50 Coffee Break  
15:50 - 16:10 Giuseppe Pirrò. Fact-checking via Path Embedding and Aggregation slides
16:10 - 16:30 Ronald Denaux and Jose Manuel Gomez-Perez. Towards Crowdsourcing Tasks for Accurate Misinformation Detection slides

Keynote Talk

WeVerify: Wider and Enhanced Verification for You

Carol Scarton

Carol Scarton

University of Sheffield, UK

twitter - website

Carolina Scarton is an Academic Fellow at the Department of Computer Science, University of Sheffield, UK. She has experience in various areas of NLP, including social media analysis, abusive language detection, machine translation, NLP evaluation and text simplification. Currently, she contributes to the WeVerify project, supporting the development of models for rumour stance classification and rumour veracity classification.


Misinformation being spread online is a great issue in our society, having a negative impact on societal debates, elections, healthcare, among others. Online misinformation can also spread fast, reaching out to millions of people. For instance, misinformation about COVID-19 has reached “misinfodemic” scale, with a worldwide spread similar to the pandemic itself. Fact-checkers are then focusing most of their effort on this topic, although it is not feasible to manually debunk every piece of misinformation. Therefore, in addition to practical verification skills, journalists and media organisations are increasingly in need of collaborative verification tools, assistance through intelligent algorithms for automatic content verification, and the ability to cross-check quickly whether a given claim or media item has already been proven false by other fact-checking organisations. In this talk I will present an overview of the WeVerify project and the tools it develops for human-in-the-loop content verification and disinformation analysis. I will focus on our work using Natural Language Processing for detecting and debunking online misinformation. Finally, our work on dealing with COVID-19 misinformation will also be presented.