Advancing UN City Resilience Efforts Using Relational Knowledge Graphs For Risk Modeling

Event Details

LocationRoom 4 (Virtual)
Date May 3, 2022
Time5:00 PM to 6:30 PM
  • Disaster preparedness has always been hard because the causes of disasters are complex and their direct and indirect impact is complex too – the behavior of humans being one of the unpredictable factors. With the world changing increasingly rapidly (technologically, climatologically and socially) the likelihood of unprecedented catastrophic events (“black swans”) is increasing.
  • The United Nations Habitat team for city resilience has developed methodologies to assess (ongoing) stresses and (sudden) shocks on cities. Currently this analysis is done mostly manually, which cannot be scaled to the tens of thousands of cities that need it. The UN Office for Information Technology has partnered with Relational.ai (RAI) to support this analysis with an automated system based on a knowledge graph. This graph is built by codifying domain knowledge from City Resilience experts and ingesting public data (satellite imagery, numerical data, policy documents, other), for a large number of indicators.  To understand how events can impact cities, we combine graph analytics and reasoning, using techniques such as neurosymbolic / probabilistic ML.  
  • The objective is to support the current UN program with a broadly accessible, open tool that allows cities to understand their vulnerabilities, simulate policies and events and prioritize investments in resilience that will improve lives.
  • Developers
  • Data scientists
  • Knowledge graph users
  • Domain SMEs – e.g., risk modeling, public policy
  • Ontologists
  • Data modelers
  • AI/Knowledge graph for good

Not required, but nice to have

  1. Knowledge graphs
  2. Ontology and taxonomy
  3. Conceptual modeling

Contact Us