Data Science Applied to Copernicus Data
Data science is a recurring and booming subject. It covers a wide panel of knowledge at the interface between statistics, Machine Learning, IT and business. Machine Learning methods, or machine learning, are characterized by the use of algorithms to solve mathematical problems from data.
Space data, and especially earth observation, is a key field of application as it provides important sets of data that need to be analysed, processed and converted into usable and tangible information. The community of young space applications developers is aware of modern processing techniques and tools, and more and more start-ups and SMEs are adopting these tools. Nevertheless, it is difficult for today’s managers that have no technical background to understand the full interest of data science in EO, its limits and how to adopt it. In addition, many applications still require a strong need of human decision and visual analysis.
In this context, the objective of this action is to propose two training sessions to these targets aiming at enhancing their level of knowledge and understanding of such disruptive technologies:
- Discovery training (one day course): for managers, project leaders, in order to help them seize the data science world and context.
- Python for Copernicus Data (three day course): technical learning of the use of Python in data science for Copernicus data, to the destination of developers and project leaders. A concrete business use case will be developed in this training.
The proposed sessions are a tailored adaptation of the following courses:
a) Introduction to Data Science (https://oslandia.com/en/trainings/ds1-introduction-to-data-science/): Presentation of some models and algorithms using Machine Learning applicate to Earth Observation interest as detection of ships for ocean monitoring purposes. And for the technical part
b) Data Science for GIS (3d; https://oslandia.com/en/trainings/ds4-data-science-for-gis/): Each training sessions will be designed and operated by CNES’s third parties, respectively leaders of training in geomatics and data science. The training will be marketed and hosted by CNES.
A strong selection of the beneficiaries will be realised through a European call for applicants.
Elected beneficiaries will have to meet the following criteria (non-exhaustive):
- Be part of a project or a company that have a top priority business using Copernicus data,
- Demonstrate a critical need for a quick and efficient capacity boosting in data science,
- Proposing a real use case for the second module
- Coming from a EU Member State or Copernicus country
Two trainings will take place, each training session bringing more or less 15 participants.
Outputs and Results
- 2 trainings sessions developed
- 30 different (2x15) participants benefiting from the training boost
- The development of concrete use cases for SMEs
- The diffusion of data science in EO sector