Computational Analysis and Visualization of Large Spatial Interaction Data, Hosted by Dr. Elizabeth Delmelle

Dr. Diansheng Guo, Department of Geography, University of South Carolina, http://www.spatialdatamining.org/, Presentation Title: Computational Analysis and Visualization of Large Spatial Interaction Data, Hosted by Dr. Elizabeth Delmelle

  • Professor Guo has a research program that focuses on developing computational methods to process and analyze huge and continuously evolving geospatial and temporal data.  His research methods help to address many applied problems related to, for example, public health, national security, forest patterns, and climate.  Dr. Guo’s research earned him a prestigous NSF CAREER grant in 2008.
  • Abstract: Spatial interaction and mobility data have become increasingly available due to the wide adoption of location-aware technologies. Examples of such data include human daily activities, vehicle movements and trajectories, and human migration, among many others. Spatial interactions (or flows) naturally form geographically embedded networks (graphs) that are usually very large and complex. For example, the county-to-county migration data in the U.S. has thousands of counties, about a million migration paths, and many variables associated with each flow and location (node). It is a challenging task to analyze and visualize such data and discover unknown information that often involve spatial pattern, graph structure, temporal trend, and multivariate relationship. This talk will elaborate on the challenges and present my latest research on the analysis and visualization of large spatial interaction data.