View the recording below, and please take a few seconds to help us make our webinars even better. Thank you!
- Presentation slides: MachineLearningGIS
- Presentation slides: 510 ADA intro NetHope
- Paper: Multi-Hazard and Spatial Transferability of a CNN for Automated Building Damage Assessment
- Netherlands Red Cross Git Hub
- A step-by-step tutorial to detect buildings
- Crowd2Map is a volunteer project that has added 5.4M buildings in rural Tanzania
- The Joint Damage Scale
Machine learning has been incorporated into spatial analysis allowing us to identify features more automatically. It also allows us to improve inferences into spatial data, knowing where we are missing pockets of people with greater significance, and identify trends. One of the benefits is that there are out of the box tools that make machine learning accessible for non-traditional data science staff. This session will focus on examples of how machine learning was used to provide impact in mapping and improve the science of 'where'.
Jacopo Margutti, Data Scientist, Netherlands Red Cross 510
Omran Najjar, AI and Advanced Data Engineer, Humanitarian OpenStreetMap Team
Kathryn M Clifton, PhD, Data Analytics and Reporting Lead, Catholic Relief Services
Bo Percival, Director of Technology Innovation, Humanitarian OpenStreetMaps Team