Nonlinear Autoregressive Exogenous ANN Algorithm-Based Predicting of COVID-19 Pandemic in Tamil Nadu

Venkateshkumar, M.; Sreedevi, A. G.; Lakshmanan, S. A.; Yogesh kumar, K. R.

COVID-19 pandemic is continuing to impact the health and life of a large population, as well as the economy of several countries worldwide. The exponential growth in the number of positive cases necessitates a prediction model for effective and judicial redistribution of available resources. This paper’s main objective is to predict the growth and spread of this pandemic to help the government take the necessary administrative decisions for healthcare preparedness and management. The research work uses artificial neural network (ANN) techniques as an efficient tool to predict COVID-19 cases. In India, Tamil Nadu is one of the severely affected states. This paper proposes a novel idea by modeling nonlinear autoregressive exogenous (NARX) ANN to predict the number of COVID-19 positive cases, discharges, and deaths soon in Tamil Nadu. The proposed prediction network is compared with feed-forward neural network (FFNN) and cascaded feed-forward neural network (CFNN), where NARX is found to achieve better accuracy with actual data. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.