Climate Change Reporting


Topic Modeling Visualizations


Czech Television Transcripts (2012-2022)


Andrea Culkova

Academy of Performing Arts in Prague

Ondrej Pekacek

Charles University

Irena Reifova

Charles University

Irene Elmerot

Stockholm University


This project investigates the discourse on climate change in Czech television, using topic modeling to identify the main themes and trends in the coverage of this issue between 2012-2022.

The following visualizations (created by D3.js & LDAVis) provide an overview of the results for different numbers of topics, using both Latent Dirichlet Allocation (LDA) and Non-Negative Matrix Factorization (NMF) techniques.


How does an AI image generation model - DALL-E 2 - imagine news media reporting on climate change?