Data Analytics for Predicting COVID-19 Cases in Top Affected Countries: Observations and Recommendations

Abstract

The outbreak of the 2019 novel coronavirus disease (COVID-19) has adversely affected many countries in the world. The unexpected large number of COVID-19 cases has disrupted the healthcare system in many countries and resulted in a shortage of bed spaces in the hospitals. Consequently, predicting the number of COVID-19 cases is imperative for governments to take appropriate actions. The number of COVID-19 cases can be accurately predicted by considering historical data of reported cases alongside some external factors that affect the spread of the virus. In the literature, most of the existing prediction methods focus only on the historical data and overlook most of the external factors. Hence, the number of COVID-19 cases is inaccurately predicted. Therefore, the main objective of this study is to simultaneously consider historical data and the external factors. This can be accomplished by adopting data analytics, which include developing a nonlinear autoregressive exogenous input (NARX) neural network-based algorithm. The viability and superiority of the developed algorithm are demonstrated by conducting experiments using data collected for top five affected countries in each continent. The results show an improved accuracy when compared with existing methods. Moreover, the experiments are extended to make future prediction for the number of patients afflicted with COVID-19 during the period from August 2020 until September 2020. By using such predictions, both the government and people in the affected countries can take appropriate measures to resume pre-epidemic activities.

Machine learning approaches in COVID-19 diagnosis, mortality, and severity risk prediction: A review

“…Research efforts for the prediction of COVID-19 can be classified into statistical and data analytics methods [ 98 ]. Experimental results for various statistical and mathematical models reveal that they provide poor predictions as they are incapable of handling large amounts of data [ 98 ].…”

“…Research efforts for the prediction of COVID-19 can be classified into statistical and data analytics methods [ 98 ]. Experimental results for various statistical and mathematical models reveal that they provide poor predictions as they are incapable of handling large amounts of data [ 98 ]. Regarding data analytics methods, the applications are mostly based on historical data only and do not consider external factors that affect the spread of the infection, such as population and median age index [ 98 ].…”

“…Experimental results for various statistical and mathematical models reveal that they provide poor predictions as they are incapable of handling large amounts of data [ 98 ]. Regarding data analytics methods, the applications are mostly based on historical data only and do not consider external factors that affect the spread of the infection, such as population and median age index [ 98 ]. To construct prediction approaches for COVID-19, ML and AI techniques have been widely used.…”

Assessment of Risk Estimates and Fatalities Involved with Covid-19

“…[7] Another interesting indicator that proved to have a significant effect on CFR was the Diet. The association between the global mortality rate of COVID-19 cases across different countries and dietary intake of different food groups was studied using an ecological study design [11]. The mortality rate is expressed in terms of CFR (Case-fatality rate).…”

“…The study suggests that nutritional factors available at country level could have a role in the mortality of COVID 19 pandemic. [11]…”