The University of Massachusetts Amherst
University of Massachusetts Amherst

Search Google Appliance


IEEE 2017 VAST Challenge

The Visual Analytics Science and Technology (VAST) Challenge is an annual contest with the goal of advancing the field of visual analytics through competition. The VAST Challenge is designed to help researchers understand how their software would be used in a novel analytic task and determine if their data transformations, visualizations, and interactions would be beneficial for particular analytic tasks. VAST Challenge problems provide researchers with realistic tasks and data sets for evaluating their software, as well as an opportunity to advance the field by solving more complex problems.


The VAST Challenge 2017 offered three mini-challenges and a grand challenge dealing with environmental problems potentially caused by human patterns of life and potentially harmful chemically laden effluent plumes being emitted from factory smokestacks. The data provided included traffic patterns, sensor data though the Boonsong Lekagul Nature Preserve, information about the Preserve, multispectral imagery and a map to help an ornithology graduate student particularly concerned with the population decrease of the Rose-Crested Blue Pipit determine who and what might be responsible.


  • Mini-Challenge 1 focused on analysis of vehicles passing through the Preserve over time.

  • Mini-Challenge 2 looked at data collected by air sampling monitors surrounding nearby factories, along with meteorological readings, to understand potential impacts smokestack effuents from factories may be having on the Pipit.

  • Mini-Challenge 3 required investigation into several months of multi-spectral imagery over the area to understand the Preserve’s general health.

  • Contestants could also integrate and synthesize their findings from the mini-challenges in a Grand Challenge to hypothesize and provide evidential support about who and what was the primary contributor to the environmental problems, and the classic why, where and how for the problem.

This year’s challenge received 58 submissions and recorded over 1100 unique downloads from 20 countries prior to the submission deadline. The VAST Challenge relies on an anonymous peer review process to provide feedback to the participants and to recommend submissions for award consideration. All submissions are reviewed by researchers in the visual analytics community. For this particular challenge, experts who work in plume analysis and multispectral image analysis were included as reviewers.


Now in its twelfth year, the VAST Challenge continues as a resource for the visual analytics research community. Through the generous support of the University of Maryland and the University of Massachusetts in maintaining the Visual Analytics Benchmark Repository [1,2], archived datasets from the past decade of VAST Challenge competitions remain freely available for use in student research projects and Visual Analytics courses worldwide. In addition, the research community is able to use these datasets along with the ground truth provided in the solution for evaluation and testing of new analytic approaches.


For more information contact Georges Grinstein<ggrinstein at> or John Fallon <jfallon at>.  Both are supported in part by Pacific Northwest National Labs.