SODAS Data Discussion 4 (Fall 2025)
Copenhagen Center for Social Data Science (SODAS) aspirers to be a resource for all students and researchers at the Faculty of Social Sciences. We therefore invite researchers across the faculty to present ongoing research projects, project applications or just a loose idea that relates to the subject of social data science.
Two researchers will present their work. The rules are simple: Short research presentations of ten minutes are followed by twenty minutes of debate. No papers will be circulated beforehand, and the presentations cannot be longer than five slides.
Discussion 1
Presenter: Yani Kartalis
Title: Content Plurality in Greece's Fragmented Media System
Abstract:
This talk presents ongoing work on measuring content similarity and diversity in large-scale news corpora. Using a comprehensive dataset of online news from Greece, I explore how (mostly) embedding-based approaches could be used to compare news content across outlets while looking at the impact of ownership concentration. The focus is on open methodological questions rather than finalized results, including validation, scalability, and interpretation of similarity measures. I especially welcome feedback on how to move from pairwise content similarity to meaningful indicators of diversity and concentration and other potential visualization-based options.
Discussion 2
Presenter: Tereza Blazkova
Title: Participatory Design in Predictive Modeling for Student Success
Abstract:
Predictive algorithms are increasingly used in decision-making across domains, yet such systems are often developed without incorporating the perspectives of those who operate them or those directly affected. This gap risks producing tools that are misaligned with real-world needs and fail to support the individuals they are meant to serve. As a result, involving stakeholders in the design and audit of AI is emerging as an important tool for mitigating adverse implications of AI on society and pursuing a more positive impact through democratizing the development. In our study, we combine participatory design with participatory AI audit to develop a dropout prediction system in higher education that reflects stakeholders’ needs. This work-in-progress presentation will outline early modelling results based on data from more than 48,000 enrollments, and will describe the planned participatory workshop and its evaluation through a survey experiment.