Congratulations to our PhD student Gill Griffiths-Jones, who has joined a project being conducted by the Alan Turing Institute in London.

Gill is currently working with the Turing Institute team as a temporary research annotator assisting the project team in developing a new training dataset for abusive content detection, based on a new taxonomy and dataset developed from the teams prior work in the area. The goal of the work is to better understand abusive forms of online content, which can inflict real damage on targeted victims and their communities and toxify public discourse. The aim is that the work will have real impact by contributing to the creation of better detection technologies, social research and policymaking.

Each week the temporary team members annotate threads from social media platforms using the developed taxonomy and during meetings with the expert team help to further refine the taxonomy and annotation methods. The taxonomy allows the annotators to determine what is and isn't hate speech and where hate speech exists to classify whether it is identity, person or affiliation-directed abuse via threatening language, derogation, dehumanisation, the use of slurs or the glorification of hateful actions, events, organisations or individuals.

The project is run by Dr. Bertie Vidgen.