UN Global Platform
Data For the World
A global collaboration to harness the power of data for better lives
Our purpose and mission
A global collaboration to harness the power of data for better lives.
We strive to enable data-driven transformation for better decision making. We seek to improve our world by providing access to trusted data for use at local, national and global levels. We will deliver a collaborative network of people and organisations, collectively building a Global Platform for trusted data, services and applications.
Under the governance of the UN Committee of Experts on Big Data and Data Science for Official Statistics (UN-CEBD) the Global Platform has built a cloud-service ecosystem to support international collaboration in the development of Official Statistics using new data sources and innovative methods and to help countries measure the Sustainable Development Goals (SDGs) to deliver the 2030 Sustainable Development Agenda.
The United Nations Global Platform was established as a collaborative environment to work together as a global statistical community and to learn together sharing knowledge, data and methods for all countries in the world.
As the platform has matured and the value it delivers has increased, the demand for access to the tools, data and methods from Task Teams around the world has increased dramatically.
UN Global Platform Regional Hubs
The UN Global Platform has 4 physical hubs all over the world working together to educate, collaborate and develop new technologies to work with new Big Data sources and methodologies.
Trusted
TRUST
Trust is central to the Global Platform. It is founded on the four pillars of: Trusted Partners, Trusted Data, Trusted Methods and; Trusted Learning through platform services.
Trust is achieved through collaboration, peer review and approval of all work undertaken on the platform.
Methods Service
Share and reuse trusted algorithms and methodologies.
The Methods service allows members of the Global Platform to find, use and publish methods and algorithms in the Cloud. For data scientists, it makes it easy to deploy algorithms and machine learning models that become API callable microservices.
Features
The Methods Service is building a library of trusted statistical methods and algorithms, facilitating international collaboration. The Cloud native infrastructure of the service and the APIs allow, for the first time, members of any international organisation to share and utilize the same algorithms.
The service:
- holds a library of algorithms – all of which are callable via API
- offers developers automation of algorithm tests, builds and deployment in a wide variety of programming languages
- supports statistical methods, AI and machine learning models
- encourages active collaboration in algorithm creation
- allows documentation and source code to be made public
- automatically scales to meet demand
- is available everywhere, all the time.
Approach
As the library of available algorithms grows, the effort needed to adopt cutting edge methodology will be substantially reduced everywhere in the world.
Active, international collaboration between data scientists in the development, review and improvement of algorithms is producing trusted methods and algorithms for production of new statistics.
Developers Service
Explore data, develop analytics and build applications securely on the Global Platform
The Developers Service gives you access to code development and data exploration environments.
Features
The Service supports users to develop new algorithms, data science pipelines and applications.
The service:
- offers Cloud-based workspaces and notebooks
- makes JupyterHub and Gitlab available over the web
- is managed and customized by UN Global Platform and UNSD teams to meet the needs of the Task Team
- supports R, Python, Julia and more languages
- is scalable to new users & intensive processing
- increases the productivity of algorithm and methods development using modern version control
- allows quick and effective collaboration in distributed teams
- offers sophisticated project management tools tailored for data scientists and data developers
- offers native container registry, supporting reproducible research and development
- automates code execution, algorithm deployment, and manages data pipelines using Gitlab runners.
Approach
This service provides environments for programmers and data scientists worldwide to collaborate on projects.