COVID 19: Data scientists have tried to answer a question that is on everyone's mind: When will the Koranic virus pandemic end?
To predict the answer they used a mathematical model also known as SIR from the initials susceptible, infected, and recovered. The mathematical model is constructed to calculate the spread and remediation of diseases.
Researchers at the University of Technology and Design in Singapore (SUTD) fed the model with data from around the world. Confirmed infections, tests performed, and deaths recorded to assess the COVID life cycle 19.
Globally, then, their system predicts that pandemic will end in December (See results for each country separately – Greece not included). However date expiration date varies greatly between states, since for example the AI states that it will expire in June at Australia or October in Italy.
The researchers say that both the model and the data can not accurately capture the complexity and dynamic nature of the disease. However, they still believe that predictions can support precautionary planning for when the lockdown may end.
History shows that the evolution of pandemics it is not completely accidental. They follow a life cycle that begins with the outbreak of the disease, accelerates to the peak of infections and then slows down until it disappears. All these stages are influenced by the government policies and individual actions of each country.
This convinced the SUTD team that data science could predict life cycle of the corona. But first, they had to collect a lot of data.
This information was found in a dataset for COVID 19 that was gathered from Our World in Data, a research organization that investigates the largest problems of the world.
The data set consists of confirmed cases and deaths collected by the European Center for Disease Prevention and Control, as well as test statistics published in official reports.
The team then fed their data to their mathematical model so that they could predict how many people would be infected in the future. The numbers appear in bar graphs that eventually form a bell curve, indicating the start, acceleration, turning point and estimated end date of the pandemic.
Researchers stress that their predictions need to be treated with caution. Excessive optimism could prematurely loosen individual discipline and government restrictions, extending the pandemic.
They warn that the accuracy of their model depends on the quality of the data, which is often unreliable and reported differently around the world.
Also, the model cannot predict the outcomes of policy decisions. For example, Singapore's move to extend the lockdown could bend the curve earlier than predicted, while the early relaxation of social distancing in USA could delay the end of the pandemic.
