Using data science to understand the COVID-19 pandemic

Abstract

Data science in pandemic The coronavirus disease, a novel severe acute respiratory syndrome (SARS COVID-19), has become a severe global health crisis due to its unpredictable nature and lack of adequate treatment. The COVID-19 pandemic has generated a strong demand for using technologies such as data science to understand or mitigate the adverse effects of the COVID-19 on public health, society and the economy (He et al., 2021). In the current era of big data, data science and data analytics have become increasingly crucial in academia, healthcare, public relationships and business operations. Machine learning (ML) models could be effective in identifying the most critical factors responsible for the overall fatalities caused by the COVID-19. However, the functional capabilities of ML models in conducting epidemiological research, especially for the COVID-19, have not been substantially explored. There are several related research methodologies regarding the COVID-19 data analytics. For instance, adopted ML models and Random Forest (RF) have been used to perform the regression modeling and provide useful information to identify the relevant critical explanatory variables and evaluate interconnections between and among the key explanatory variables and the COVID-19 case and death counts (Gupta et al., 2021). Time-series analyses have been used to examine the rate of incidences of the COVID-19 cases and deaths (Khayyat et al., 2021). Social network analysis (SNA) has been used to track cases and simulations for modeling the COVID-19 outbreaks (Bahja and Safdar, 2020). Researchers have built models to interpret patterns of public sentiment on disseminating health-related information and assess the political and economic influence of the pandemic.