The sudden emergence of a Coronavirus and its rapid spread due to the globalization factors, especially the airline network, provoked the reaction of countries. Governments attempt to use all available means, including prediction methods, to control the spread of the Co
More
The sudden emergence of a Coronavirus and its rapid spread due to the globalization factors, especially the airline network, provoked the reaction of countries. Governments attempt to use all available means, including prediction methods, to control the spread of the Coronavirus. In this article, we have developed various models based on artificial neural networks, including multi-layer perceptron, radial basis function, and adaptive-network-based fuzzy inference system with different learning algorithms, transfer functions, membership functions, hidden layers, hidden neurons, and kernels. We have identified five factors influencing the Coronavirus outbreak based on the Pearson correlation coefficient approach. These factors are gasoline consumption, internet pressure, number of wedding ceremonies, online transactions, and mask consumption. The accuracy of the developed models is identified by calculating three types of statistical errors, including root mean square error, mean absolute error, and mean absolute percentage error. The results show that the radial basis function model predicts the number of Covid-19 cases for the one month (mid-term) with an accuracy of over 97%. This study provides an efficient approach to predict the number of COVID-19 cases which help policymakers to make strategic decisions, including closing borders, designing supply chains for medical and health equipment, and enacting new laws.
Manuscript profile
