KnowWhereGraph: Enriching and Linking Cross-Domain Knowledge Graphs using Spatially-Explicit AI Technologies
Knowledge graphs are not merely a set of technologies but a novel paradigm for representing, retrieving, and integrating data from highly heterogeneous sources. Knowledge graphs have become a core component of modern search engines, intelligent personal assistants, business intelligence, supply chain management, and many other application areas within just a few years. KnowWhereGraph aims at adding environmental intelligence to this rapidly growing list of application areas. Our graph consists of over 13 billion triples (graph statements) across 30+ different data layers. These layers consist of spatial identifiers such as places, regions, hierarchical discrete global grid cells, and so on, as well as thematic layers on top of them including population health, past and present extreme events, transportation infrastructure, expert networks, and agricultural data about soils and crops. Put differently, for every spatial identifier such as a ZIP code, we provide access to thematic data to enrich a decision-maker’s own data with further contextual information.
KnowWhereGraph is the first to fuse a knowledge graph with geo-enrichment, thereby making the graph directly available from within Geographic Information Systems (GIS) and other environments. This also includes bespoke solutions such as services and interfaces for humanitarian relief, food supply chains, credit risk, and so on.
The KnowWhereGraph Ontology can be found here. A series of developed tools and their use cases can be found here.