Clemens Baldzuhn

Researcher and PhD-Candidate @GESIS

Welcome to my website! This website serves to give an overview of the academic and non-academic projects I have pursued in the past.

I’m a researcher and doctoral candidate within the Designed Digital Data team in the Computational Social Science department at GESIS. As part of the DFG-funded NewsFrag project, I investigate the dynamics of news audience fragmentation by integrating web-tracking and survey data.

Methodologically, I employ Bayesian modelling (e.g. for point and Gaussian processes) and information-theoretic approaches to capture social dynamics.

I hold a M.Sc. in Statistics from the Berlin University Alliance, as well as a B.A. in Sociology and a B.Sc. in Psychology from the University of Leipzig and Maastricht University.

Projects

As a part-time project, we developed an automatic fact-checking tool for German talkshows using agentic AI (ongoing, demo available).

For the Tarmac Festival, I developed a web app–based ticketing system to streamline on-site transportation requests.

I developed a novel Bayesian informed statistical model for estimating latent attitudes from likert scales in panel data (ongoing).

For my Master’s thesis in Statistics I developed new statistical methodology to describe a special type of point data (Currently in publication).

During an internship at the Complexity Science Hub I worked with statistical copulas and information theoretic metrics to extend a model of armed conflict spread.

As an improvement to my workflow finding new records on the web, I came up with a Streamlit powered dashboard, which uses the Discogs API to display the most relevant statistics of a record label.

I developed a Telegram Chatbot that offers the nearest free public toilet in Berlin and Dresden, which had several hundred users when it was still online.

For my Bachelor’s thesis in Leipzig I conducted a large-scale data analysis of job advertisements from the German labour market.

During my time in the Computational Social Science department at Gesis, I developed a pipeline to analyze deleted tweets of politicians in their extensive collection, which was gathered during the German federal elections in 2017 (BTW17).

Analyzing tweets from German police stations (Oct. 2020 - Feb. 2021). This is a data science project for the course 'Introduction to Digital Humanities' at Leipzig University.

© Clemens Baldzuhn 2026