The project, led by SnT vice-director professor Yves Le Traon, is unique, says the university, because it delivers a method to model the impact of different measures on the spread of covid-19 in a large number of countries around the world.
“The tool uses machine learning techniques to analyse public data and deliver hypothetical projections of how different isolation measures will impact the spread of covid-19. The intention is to make it possible for experts and governments around the world to analyse how various exit strategies will impact the spread of covid-19 in a six-month timeframe.”
The modelling tool is using workplaces, outdoor activities, public transportation, and retail as its four variables. It uses data that is publicly available from the Google covid-19 dataset.
The researchers are releasing a beta version of the tool in order to benefit the public as soon as possible. “As many countries in Europe are beginning to execute on their plans already, we wanted to release our work as soon as possible,” said Le Traon.
Additionally, the teams had arrived at the point in the development process that required the feedback of users in order to refine the machine-learning algorithm that drives it.
“The simulator is a ground-breaking instrument with the potential to enhance the covid-19 exit strategy planning of all included countries,” said University of Luxembourg rector Stephane Pallage. “Covid-19 requires agile solutions and professor Le Traon’s team have reacted swiftly by disclosing a valuable technology that allows to model possible effects of public policy decisions as exit strategies are being planned and implemented.”