The European Space Agency developed a Machine Learning (ML) application, called Sen2Agri, which recognizes patterns of pixels from satellite imagery as specific types of agricultural crops. Using Sen2Agri, we have built a data pipeline, to process every month satellite imageries covering the whole of Senegal to determine the state of all its crop statistics. Similarly, shapes of inland water surfaces, ships (or types of ships) and conditions of roads can be recognized by ML programs. Such applications are very useful to monitor the extent of freshwater resources, the occurrence of illegal fishing or the access to all-season roads for rural populations.
Machine Learning for pixel recognition is an example of Data Science, which is broadly defined as applying the tools, methods and practices of the digital and data age to create new understanding and improve decision-making. In such a way, Data Science can help the community of Official Statistics to better inform the Government, businesses, the research community and the public in general about the economic, demographic, social and environmental situation.
At the beginning of this year FSO established a Data Science Competence Centre which will benefit all government departments of Switzerland, while UNSD has been supporting the UN Committee of Experts on Big Data and Data Science for Official Statistics since 2014. Facebook uses Data Science in developing market strategy, but also to support a large Open Source community, called PyTorch, which has developed a machine learning framework that accelerates the path from research prototyping to production deployment.
The Federal Statistical Office (FSO) of Switzerland together with the United Nations Statistics Division (UNSD) and Facebook are organizing a webinar on the topic of Data Science and Official Statistics. The webinar is part of the “Road to Bern” events and will take place on 30 June 2021 from 16:00 PM to 17:30 PM, time in Bern, Switzerland (10:00 AM to 11:30 AM, New York time).
The program of webinar:
Introduction – Ronald Jansen, Chief Data Innovation and Capacity Branch, United Nations Statistics Division
Machine Learning framework – Jyothi Nookula, Facebook AI
Machine Learning and Official Statistics – Oliver Mahoney, Data Science Campus, Office for National Statistics, UK
Data Science throughout the Swiss government – Prof. Dr. Bertrand Loison, Vice Director, Federal Statistical Office, Switzerland
Remote Data Science: working with data you don’t see – Andrew Trask, Leader, Openmined
Q&A Panel discussion
Registration link: https://forms.gle/QLv9Xb8H3Sy7BotC6