Understanding the Covid-19 pandemic: A machine learning approach

Rediet Mengistu

Master's thesis

Abstract:

Since the onset of the Covid-19 pandemic, scientists have been working frantically to learn more about the coronavirus's behaviour in order to keep it from spreading further. That's where big data comes in, delivering much-needed data to aid in the battle against the pandemic. This diploma thesis seeks to assess the covid-19 pandemic by using machine learning techniques with the goal of adding to the much-needed information about the virus. This thesis tackles the first objective of clustering countries based on spread pattern by using hierarchical clustering algorithm. Five distinct clusters of countries of different sizes are uncovered when assessing the spread pattern. These clusters showed different patterns of the spread of the virus in different time segments. This study also investigates how the pandemic impacted the clusters from an economic education and health standpoint after grouping countries together based on the virus's distribution pattern. Growth rate and unemployment rate are used as metrics to assess the economic effects of the pandemic. Both metrics suggest that the pandemic has had a negative effect on the economy of the clusters. To investigate the effect of the pandemic on education, school closures in each cluster for the year 2020 were analyzed and linked to the availability of internet in that cluster to show how students without internet were affected. Finally, health impact of Covid-19 are investigated, demonstrating how hospitals found in each cluster were overburdened because of the pandemic. As the coronavirus advanced, governments took a variety of approaches to dealing with the outbreak with the intention of influencing the virus's trajectory. Hierarchical clustering is also implemented here based on the stringency of the government response and 3 distinct clusters were found. The clusters are further investigated with respected to individual policies member countries have followed.

Keywords

Big Data, Python, Coronavirus, New Cases, Stringency Index

Language used: English

Date on which the thesis was submitted / produced: 30. 3. 2021

Thesis defence

• Supervisor: Ing. Tomáš Hlavsa, Ph.D.

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme / field:
Systems Engineering and Informatics / Informatics