Invited talks

  • Detecting narration changes in economics utilizing continuous topic modeling and (large) language models. Research Colloquium of the Faculty of Business Studies and Economics, JLU. Gießen, Germany (01/2025).
  • Monitoring (social) media narratives combining retrospective few-shot classification with continuous topic modeling. CMStatistics 2024. London, England (12/2024).
  • PETapter: A masked-language-modeling classification head for modular fine-tuning of (large) language models. CompStat 2024. Gießen, Germany (08/2024).
  • Bekämpfung von Desinformation und Fake-News durch GADMO. BDI/BDA-Arbeitskreis Statistik. Düsseldorf, Germany (05/2024).
  • Keep rollin’! The abilities for monitoring growing corpora using RollingLDA. ZPID Lecture Series. Trier, Germany (12/2022). Recording.

Contributions to conferences and workshops

  • Exploring the potential of large language models (such as GPT-4) for (semi-)automatic content analysis of stances and frames in media texts. DGPuK 2024. Erfurt, Germany (03/2024). Abstract.
  • Few-shot learning for automated content analysis (FLACA) in the German media debate on arms deliveries to Ukraine. Digital Total. Hamburg, Germany (10/2023). Poster.
  • Debunking Disinformation with GADMO: A Topic Modeling Analysis of a Comprehensive Corpus of German-language Fact-Checks. DiTox’23 Workshop @LDK 2023. Vienna, Austria (09/2023).
  • Bekämpfung von Desinformation durch GADMO: Analyse eines umfassenden deutschsprachigen Faktencheck-Korpus mithilfe von Topic Modellen. Statistische Woche 2023. Dortmund, Germany (09/2023).
  • Scrutinizing ChatGPT against Few-Shot Learning with Adapter Extensions and XLM-RoBERTa: A Case Study on Identifying Claims, Arguments and their Stance in the German News Media Debate on Arms Deliveries to Ukraine. Statistische Woche 2023. Dortmund, Germany (09/2023).
  • Beyond “Master Frames”: A Semi-automated Approach to Studying Viewpoint Diversity of the Media Discourse. ECREA PolComm 2023. Berlin, Germany (08/2023).
  • Few-shot learning for automated content analysis: Efficient coding of arguments and claims in the debate on arms deliveries to Ukraine. DGPuK 2023. Bremen, Germany (05/2023).
  • Early Warning Systems? Building Time Consistent Perception Indicators for Economic Uncertainty and Inflation Using Efficient Dynamic Modeling. MUFin’23 Workshop @AAAI 2023. Washington, DC, USA (02/2023). Pre-recording.
  • Finding scientific topics in contionuously growing text corpora. SDP’22 Workshop @COLING 2022. Gyeongju, Republic of Korea (10/2022). Pre-recording.
  • Monitoring consistent topics in continuously growing scientific text corpora. Statistische Woche 2022. Münster, Germany (09/2022).
  • Dynamic change detection in topics based on rolling LDAs. Text2Story’22 Workshop @ECIR 2022. Stavanger, Norway (04/2022). Pre-recording.
  • Improving the reliability of LDA results using LDAPrototype as selection criterion. DAGStat 2022. Hamburg, Germany (03/2022).
  • RollingLDA: An Update Algorithm of Latent Dirichlet Allocation to Construct Consistent Time Series from Textual Data. EMNLP 2021. Punta Cana, Dominican Republic (11/2021). Pre-recording.
  • Assessing the Uncertainty of the Text Generating Process using Topic Models. EDML’20 Workshop @ECML PKDD 2020. Online (09/2020).
  • Improving Latent Dirichlet Allocation: On Reliability of the Novel Method LDAPrototype. NLDB 2020. Online (06/2020).
  • Quantifizierung der Stabilität der Latent Dirichlet Allocation mithilfe von Clustering auf wiederholten Durchläufen. Statistische Woche 2019. Trier, Germany (09/2019).
  • Softwaretools für die Kommunikationsforschung. DGPuK 2019. Münster, Germany (05/2019).
  • Measuring Stability of Replicated LDA Runs. DAGStat 2019. Munich, Germany (03/2019).