This project is one of the Scrum projects with industry partners that were part of the AMOS Summer 2026 Projects. Below please find the video (you may also like the other videos) and the project summary which details the final result of the project. We run these projects every semester, so please be in touch if you would like to motivate one of your own!
Demo Video
Project Summary
| Project name | Automated Thematic Analysis |
| Project mission | To engineer a secure, open-source tool that ingests thousands of subject interviews about a policy topic and produces an accurate thematic analysis breakdown. The core is a multi-stage LLM pipeline that extracts quotes, codes them into themes, clusters and ranks those themes, and cross-references them against respondent demographic dimensions. Results are surfaced through a lightweight frontend and exportable in structured formats. We also provide direct reference back to the corpus, so researchers can see which interviews most strongly manifest a particular theme or demographic combination. |
| Industry partner | Nürnberg Institut für Marktentscheidungen |
| Team logo |
|
| Project summary | The product supports social science researchers in efficiently analysing large interview corpora. It combines Large Language Models (LLMs) with modern machine-learning algorithms to systematically structure, analyse and clearly visualise qualitative data. First, researchers import interview transcripts along with demographic information about participants into the software. Based on this data, the system automatically generates a codebook: a hierarchical topic structure that summarises the key themes across the entire interview corpus. In the next step, an LLM applies the identified topics to individual interviews and specific passages of text. This makes it possible to see which topics occur in which interviews and where they are discussed. The results can then be exported or explored directly within the application. In particular, researchers can visualise thematic differences across demographic dimensions. For example, they can examine whether, and how frequently, specific groups of participants address a particular topic. The product accelerates qualitative research analysis, creates transparency throughout the analytical process and makes comparative analysis across large volumes of interview data significantly easier. |
| Project illustration |    |
| Team photo |
|
| Project repository | https://github.com/amosproj/amos2026ss02-automated-thematic-analysis |
Discover more from Professorship for Open-Source Software
Subscribe to get the latest posts sent to your email.