ANFIS Algorithm-based Modeling and Forecasting of the COVID-19 Epidemic: A Case Study in Tamil Nadu, India

Abstract

COVID-19 manifests itself as a transferable disease caused by the newly-exposed coronavirus, which originated in Wuhan province, China, in December 2019. COVID-19 swiftly spread through almost the entire world including India. Due to the lack of an effective vaccine and lack of awareness among people, the global population has been hit hard by the COVID-19 outbreak and invaluable human lives have been lost. In this chapter, efficient mathematical models using Adaptive Neuro Fuzzy Inference System (ANFIS) are recommended for forecasting COVID-19 infection and fatality rates, together with the number of active cases at any given point in time. The models were developed using available data on the spread mechanism of COVID-19 in the State of Tamil Nadu, India. By utilizing official data concerning reported cases of COVID-19 provided by the State Government of Tamil Nadu, experiments were performed using the developed ANFIS models. The results signify that these epidemic models can forecast various forms of COVID-19 cases, infection and death rates, and the number of active cases with a fair degree of accuracy. A close match between predicted results and actual outcomes shows the accuracy of the proposed ANFIS models for the COVID-19 pandemic.