My research is characterized by an interdisciplinary perspective on geographic information science driven by advances of spatial Big Data (e.g., social media and remote sensing), machine learning/artificial intelligence, and computational sciences. I am interested in deep learning of heterogeneous geographic information to support uncertainty-aware geographic knowledge discovery and decision making. Particularly, I focus on the development of novel statistical/machine learning and computational methodologies for (a) integrating heterogeneous sources of geographic information (e.g., incompatible scales) for geographic analysis; (b) characterizing and modeling complex spatiotemporal patterns (e.g., land cover and land use changes); (c) characterizing and modeling spatiotemporal bias and uncertainty of geographic information and the associated impacts; (d) addressing the computing challenges when the data scale and model complexity are not manageable; and (e) building geospatial-enabled cyberinfrastructure for domain scientists. Domain applications focus on natural hazards, environmental sciences, public health and global changes.
Recent Courses Taught
- Spring 2021 GEOG 4023/5023 Quantitative Methods
- Spring 2021 GEOG 4103/5103 GIS: Spatial Analytics
- Fall 2020 GEOG 3023: Statistics for Geography