Spatio-temporal grounding of claims made on the web

Social media presents us with a digitally-accessible sample of all human discourse. This sample is full of claims and assertions. While the state of the art in NLP is adapting to the volume, velocity and variety of this sample and the information in it, the accuracy of claims made in social media remain largely unstudied. PHEME, a 36-month EU project started in January 2014, focuses on this fourth challenge:veracity. As a core part of establishing veracity, we need to identify the spatio-temporal context of assertions made on informal websites. This project note introduces the spatio-temporal challenges and planned semantic annotation activities that are part of the PHEME project.