Seminar Data-Science III (Social Data Science)

The achievement of the learning goals is pursued by practicing on the basis of personally assigned in-depth scientific topics as well as by actively participating in the presentation dates. The organizer will choose subject areas within the field of Data-Science (see Topics) and provide scientific papers to students to work through.

Previous participation in the courses offered by our chair are recommended.

This seminar will be split into two main topic blocks. Every student will be assigned a research paper from only one of these blocks to work on. Yet, it is expected that students also actively participate in discussion on papers from the other topic blocks after they have been presented.

When applying for this seminar, please indicate whether you would be interested in only one or both topic blocks. The two topics we are going to discuss in the FSS 2026 are:

1. Cultural and Creative Homogenization in LLMs. As LLMs are increasingly used for writing and idea generation, they tend to favor familiar, dominant styles, shaping how people express themselves and narrowing the range of voices and perspectives. We will examine how implicit assumptions, alignment practices, and human preferences drive homogenization and how these forces can be measured. Finally,  we will explore strategies and emerging methods to improve diversity, preserving cultural nuance and expanding creative variation in human–AI co-creation.
   • AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuances
   • Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond)
   • Improving Diversity of Demographic Representation in Large Language Models via Collective-Critiques and Self-Voting
   • Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity
2. Predictive Privacy. Modern machine learning often focuses on predicting a person's private details and future actions based on the behavior of others. In this seminar, will discuss Predictive Privacy, an ethical framework designed to protect against differential treatment fueled by big data. The core challenge lies in the prediction gap, where statistical inferences derived from broad cohorts are transformed into specific, invasive individual profiles. Keeping our information safe is no longer just a personal task, but a shared responsibility in a world where everyone's data is linked.
   • Predictive privacy: towards an applied ethics of data analytics
   • Contextual Integrity Up and Down the Data Food Chain
   • From Individual to Group Privacy in Big Data Analytics
   • Privacy as a Common Good in the Digital World

Through this seminar, students will gain a comprehensive understanding of the ethical and social dimensions of machine learning and AI, preparing them to critically engage with these technologies in their future work.

On the basis of suitable literature, in particular original scientific articles, students independently familiarize themselves with a topic in data-science, classify and narrow down the topic appropriately and develop a critical evaluation. Students work out concepts, procedures and results of a given topic clearly and with appropriate formalisms in a timely manner and to a defined extent in depth in writing; Evidence of independent development by presenting self-selected examples. Descriptive oral presentation of an in-depth data science topic using suitable media and examples in a given format.

To be announced.

If you are interested in this seminar, please apply to Marlene Lutz via email. 

Please provide some details about your background, e.g., whether you have taken some relevant classes before and a short motivation to take this seminar. Also, make sure to indicate which ones of the four given topics you are interested in.