Bar charts are an effective way to convey numeric information, but today's algorithms cannot parse them. Existing methods fail when faced with even minor variations in appearance. Here, we present DVQA, a dataset that tests many aspects of bar chart understanding in a question answering framework. Unlike visual question answering (VQA), DVQA requires processing words and answers that are unique to a particular bar chart. State-of-the-art VQA algorithms perform poorly on DVQA, and we propose two strong baselines that perform considerably better. Our work will enable algorithms to automatically extract numeric and semantic information from vast quantities of bar charts found in scientific publications, Internet articles, business reports, and many other areas.
DVQA: Understanding Data Visualizations via Question Answering
DVQA, a bar chart understanding dataset, tests the ability of algorithms to parse bar charts through question answering, addressing shortcomings of current VQA methods and enabling automatic extraction of numeric and semantic information.
- Year
- 2018
- Venue
- dvqa-understanding-data-visualizations-via-1
- Authors
- 4
- Hosting
- Abstract onlyARXIV-DEFAULT
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/1801.08163v2ARXIV-DEFAULT
- TL;DR
- Semantic Scholar