Collecting and Characterizing Natural Language Utterances for Specifying Data Visualizations
arXiv:2110.00680 · doi:10.1145/3411764.3445400
Abstract
Natural language interfaces (NLIs) for data visualization are becoming increasingly popular both in academic research and in commercial software. Yet, there is a lack of empirical understanding of how people specify visualizations through natural language. To bridge this gap, we conducted an online study with 102 participants. We showed participants a series of ten visualizations for a given dataset and asked them to provide utterances they would pose to generate the displayed charts. The curated list of utterances generated from the study is provided below. This corpus of utterances can be used to evaluate existing NLIs for data visualization as well as for creating new systems and models to generate visualizations from natural language utterances.
Paper appeared at the 2021 ACM Conference on Conference on Human Factors in Computing Systems (CHI 2021), 10 pages (5 figures, 3 tables)
References in corpus (3)
Cited by in corpus (11)
- Towards Natural Language Interfaces for Data Visualization: A Survey
- ChartGPT: Leveraging LLMs to Generate Charts from Abstract Natural Language
- Towards Natural Language-Based Visualization Authoring
- The Reality of the Situation: A Survey of Situated Analytics
- CrossData: Leveraging Text-Data Connections for Authoring Data Documents
- XNLI: Explaining and Diagnosing NLI-based Visual Data Analysis
- Sporthesia: Augmenting Sports Videos Using Natural Language
- What Does the Chart Say? Grouping Cues Guide Viewer Comparisons and Conclusions in Bar Charts
- Natural Language Dataset Generation Framework for Visualizations Powered by Large Language Models
- Epigraphics: Message-Driven Infographics Authoring
- Chart What I Say: Exploring Cross-Modality Prompt Alignment in AI-Assisted Chart Authoring