ABSTRACT - Data are often imperfect despite the advances in data collection technologies. How to incorporate such uncertainty into data visualization has been challenging. This research discusses the use of choropleth mapping to visualize geographical data that contain uncertainty. A new map classification scheme is introduced to account for a wide range of data uncertainty. To address the possibility that a geographical unit might be placed in a wrong map class due to data uncertainty, a robustness measure is proposed and integrated into the optimal design of choropleth maps. Solution algorithms are developed along with a novel theoretical bound for evaluating solution quality. The new approach is applied to map the American Community Survey data. The study provides an important perspective on addressing data uncertainty in map design and offers a new approach for spatial data analysts to incorporate robustness into data visualization.
BIO - Dr. Daoqin Tong is an Associate Professor in the School of Geographical Sciences and Urban Planning at Arizona State University. Dr. Tong received her M.S. in Civil Engineering, M.A.S. in Statistics, and Ph.D. in Geography from the Ohio State University. Dr. Tong’s research has primarily focused on spatial data analytics, geographic information science, spatial optimization, spatial statistics and big data to support urban system design and operation with applications in facility location, transportation, urban sustainability, and public health.