Nathan McEachen

Attendees will build two complementary approaches for enabling geospatial awareness in LLMs: Geo-GraphRAGs for analyzing networks of features and DGGS-based methods for integrating aggregate and statistical data, and how they work together.

While LLMs excel at synthesizing text, they lack geospatial awareness and cannot reason over spatial networks. This talk demonstrates how Geo-GraphRAG and DGGS, grounded in open standards, enable geospatial reasoning with transparent, traceable results.

Attendees will build two complementary approaches for enabling geospatial awareness in LLMs: Geo-GraphRAGs for analyzing networks of features and DGGS-based methods for integrating aggregate and statistical data, and how they work together.

While LLMs excel at synthesizing text, they lack geospatial awareness and cannot reason over spatial networks. This talk demonstrates how Geo-GraphRAG and DGGS, grounded in open standards, enable geospatial reasoning with transparent, traceable results.

Graph-based retrieval-augmented generation (GraphRAG) can capture the semantics of geoinformatics data, which can be leveraged by LLMs. We will demonstrate how to build a GraphRAG and how to write prompts so that an LLM Agent can utilize it.

National spatial data infrastructures (NSDIs) contain a wealth of information that AI-based solutions can leverage to improve operational efficiency. We propose a collaborative strategy to bring geospatial awareness into LLMs at scale by constructing GraphRAGs for LLM Agents.

Geospatial awareness can be brought to LLMs to address humanity’s most pressing challenges by using authoritative and cross-sectoral Spatial Knowledge Graphs that are automatically interoperable by location. We will present a technical approach using open-source software.

KnowSTAC is a framework for creating dynamic STAC indexes that organize data not according to how they are published as collections or catalogs, but according to on-demand queries that can crosscut domains and organizations for improved collaboration.