NCSI: Geospatial Analysis Through Graph Learning (Intern Position)

Status	Completed
Mentor Name	Xinlian Liu
Mentor's XSEDE Affiliation	Campus Champion, Education Allocation
Mentor Has Been in XSEDE Community	4-5 years
Project Title	Geospatial Analysis Through Graph Learning (Intern Position)
Summary	Geo-spatial analysis is challenging because samples are not independent, but rather under influence by others in neighboring regions. Statistics-based methods of geospatial analysis rely on assumptions of adjacent interference, which is hard to characterize. We will explore using Graph Learning to detect geospatial clusters.
Job Description	1. Collect county-level historic data from various sources, such as ACS, CDC, etc. 2. Tune deep learning networks for community characterization 3. Perform literature review and write technical documents
Computational Resources	TACC Stampede2
Contribution to Community	1) Validate an innovative approach in geosptail analysis 2) Create a workflow and code repository for such task
Position Type	Intern
Training Plan	We will perform extensive training on using XSEDE resources, programming with PyTorch, etc.
Student Prerequisites/Conditions/Qualifications
Duration	Semester
Start Date	05/15/2021
End Date	07/31/2021