This month, BTAA-GIN is highlighting a new step-by-step tutorial that shows students and faculty how to apply deep learning models to geospatial data. The tutorial uses high-resolution aerial imagery from the BTAA Geoportal to teach machine learning workflows for spatial analysis.
What makes this possible? BTAA-GIN’s work curating geospatial data and applying standardized workflows and metadata. This behind-the-scenes effort to ensure authoritative, well-described data means researchers can move directly from discovery to hands-on analysis using familiar GIS and data science tools to explore patterns, detect features, and ask new spatial questions.
As BTAA-GIN continues shaping its FY26–FY28 roadmap, examples like this show how shared collections continue to evolve alongside changing research and teaching practices across Big Ten universities.
Note: We’re unable to publish our usual search analytics this month. We’re investigating an issue where Google appears to have indexed one of our internal metadata pages as if it were a dataset, which has skewed our search traffic data. We’re working to resolve this and will resume reporting accurate search trends once the issue is corrected.
Presented to the Geo4Lib community on the BTAA Geospatial API and the future of OpenGeoMetadata, with a focus on scalable onboarding for new institutions.
Streamlined the platform by unifying the backend and frontend into a single codebase and migrating to server-side rendering to improve performance and security.
Improved the user experience and infrastructure with new search views, institutional branding, IIIF viewing, performance optimizations, and support for very large file uploads.
