Data science and artificial intelligence
Gaining critical insights on vulnerable populations faster in emergencies

Exogenous shocks such as population displacements, disease outbreaks, and natural disasters are sudden and often unpredictable.
Data can provide the global community, governments,and international organizations like UNICEF with new and faster ways of analysing the world that generate critical insights into the needs of vulnerable populations, and where best to invest our scarce resources in emergency response.
The COVID-19 pandemic poses an extraordinary challenge to the world, our societies, health care systems, and economies.
We are continuously studying and predicting the effects of mobility on COVID-19 response with real-time data.
Magic Box: a collaborative data sharing platform
UNICEF is investing in Magic Box, a collaborative platform that is made possible through the contributions of private sector partners such as Telefonica, Google, IBM, Amadeus and Red Hat which share their data and expertise for public good. By harnessing real-time data generated by the private sector, UNICEF can gain critical insights into the needs of vulnerable populations, and make more informed decisions about how to invest its resources to respond to disaster, epidemics and other challenges. The Magic Box creates a provocation to industry, a platform for engagement and a product that can change the way UNICEF works in emergencies like Zika or Ebola.
Key Applications
How it works
Connecting real time data generated by private sector to the humanitarian response tools is a challenging task that has never been done before. It requires new type of infrastructure to allow these distant ecosystems to connect. With the help of partners, UNICEF’s Office of Innovation is developing a software platform that intended to use real-time data to inform life-saving humanitarian responses to emergency situations. This open-source platform ingests data from both public sources and from private sector partners, and generates insights based on methodologies and algorithms provided by our data science team. These insights are made available to the development and humanitarian partners through an API and user interfaces.
The first version of this platform was created during the 2014 Ebola crisis in West Africa, and a second version developed with Google in response to the Zika outbreak in 2015. Since then, it is being adapted to multiple applications and made available to open-source collaborators.
There are several solutions that address the use of a particular dataset and also solutions that address the problem but under proprietary software. The main differentiators of our solution are:
- Combination of different datasets from diverse providers for a more insightful result.
- Proactiveness. Many initiatives requiring data sharing are only established at the time of a crisis. Our approach is to partner with data providers in advance, to allow preparedness activities and ensure readiness at the time of an emergency
- Open source software and open methods published
- Partnership with academia and cutting-edge data scientists to develop the solution
- Our multidisciplinary team, composed by designers, engineers and scientists, complement UNICEF 70-year field expertise in order to tailor a product that fits real needs.
Stories and Research
Partners
Get Involved
Magic Box is a collaborative platform, that can only be made possible by contributions of multiple partners that bring their data and expertise for public good. UNICEF’s Office of Innovation is actively looking for partners and contributors to build on our systems for real-time information collection, aggregation, and analysis.