AI/Machine Learning-based system capable of predicting forced displacement in near real time
THE QUESTION
Is it possible to develop an Artificial Intelligence-based system to predict forced displacement in near real time?
LOCATION: Ukraine
SECTOR: Humanitarian
TECH: AI
TIMELINE: August 2023 - Present
PIONEERS: Dan Caspersz, Liam Avnon, Noam Rosenbaum
PARTNERS: Save the Children, Brunel University
The Challenge
We know that more timely and more accurate data on forced displacement has the potential to transform the way humanitarian aid is provided. This pilot will explore the use of AI and Machine Learning-based models applied to live data sources and proxies for behaviour to predict displacement. With this information, the team is looking to explore how the data and insight produced could be used by the UK government and its partners to improve the targeting of humanitarian aid and deployment of staff.
The Idea
In partnership with the Frontier Tech Hub, FCDO are seeking to develop an AI and machine learning-based system to predict forced displacement with high levels of accuracy in near real-time. Beyond developing the system, FCDO would like to explore how predictions can be used by the UK government and its partners to improve the targeting of humanitarian aid and deployment of staff. The system is expected to be trained with historical displacement data and then combine this with digital signals data to predict where displacement is likely to take place. The goal is to develop a proof-of-concept model in Ukraine, that could then be scaled up and applied to other geographies.
Our learnings and stories so far
This pilot hasn’t started to publish yet, but there are plenty of other blogs to read below. Check back soon!
The Frontier Tech Hub is seeking technical support in land rights for smallholder farmers in Colombia. The consultant will use their deep expertise to help analyse challenges, explore root causes, and assess market and landscape factors affecting land tenure.