Ströbel, MaraEckert, KaiNagel, TillKucher, KostiantynDiehl, AlexandraGillmann, Christina2024-05-212024-05-212024978-3-03868-258-5https://doi.org/10.2312/evp.20241083https://diglib.eg.org/handle/10.2312/evp20241083This paper explores the effectiveness of an LLM in creating data visualizations across a spectrum of scenarios, characterized by three key dimensions: the complexity of the underlying data, the user's data visualization competencies, and the requirements of the resulting visualization. Based on an empirical study, we offer insights into the potential role of LLMs as tools for empowering users with varied expertise to effectively visualize data.Attribution 4.0 International LicenseCCS Concepts: Human-centered computing → Visualization; Computing methodologies → Natural language processingHuman centered computing → VisualizationComputing methodologies → Natural language processing10.2312/evp.202410833 pages