Utrecht University AI Lab Digital Humanities funding awarded to Data School for a co-financed PhD position with De Groene Amsterdammer
Principal investigator: Karin van Es (Data School)
Co-applications: Mirko Tobias Schäfer and Joris Veerbeek (both Data School), Kees van Deemter (Information and Computing Sciences), Ayoub Bagheri (Methodology and Statistics), Beatrice de Graaf and Pim Huijnen (both History);
PhD ‘Employing text classification models for journalistic inquiry on public debates‘ – Joris Veerbeek, MA
Together with Dutch journalistic weekly De Groene Amsterdammer, the aim of this project is to investigate how state-of-the-art language models can be fruitfully applied for journalistic inquiry. Shaped into a PhD position, it will (a) develop and test strategies for using automatic text classification in a journalistic setting, and (b) use text classification models to study public debates and the dynamics of social media.
This project will focus on three (use) cases:
- Toxicity and hate-speech detection
- Stance detection
- Contextual embeddings and language representation
The first case aims to develop a generalised toxicity detection model that will be evaluated across different social media platforms to study out-group aggressiveness and its development over time. The second case aims to develop and evaluate a framework to model stances within texts without having to adapt the training data to the specific topic of interest. The final study evaluates the use of contextual embeddings to study the interaction between discourse and interactions on social media.
De Groene Amsterdammer will use the results and the models developed within the context of this PhD project for a series of articles within the branch “Data & Debate”.
In the AI Labs at Utrecht University, academic researchers collaborate with external partners on societal issues that involve artificial intelligence and/or data science. With the AI Labs, UU encourages collaborations between researchers from the faculties of Science, Humanities and Geosciences and others, with external partners. The three AI Lab-themes for this are AI & Sustainability, AI & Life Sciences and AI & Digital Humanities.