A Comparison Between Adaptive Neuro-fuzzy Inference System and Autoregressive Integrated Moving Average in Predicting COVID-19 Confirmed Cases in Bangladesh

Abstract

Since December 2019, the novel coronavirus (COVID-19) has become one of the most contagious diseases to have hit the world for several decades. From December 2019 till May 2020, this respiratory syndrome-like disease has quickly spread to all countries around the world and has taken more than 400 thousand lives. The WHO declared a global pandemic situation due to the virus from March 2020. The source of this virus is not known, especially since there are no well-placed standards for its diagnosis and treatment. Several factors are involved in the spread of the disease. There have been several studies to predict or forecast the number of new cases in upcoming dates. In our study, we tested the widely used ANFIS—Adaptive Neuro-Fuzzy Inference System and the ARIMA—Autoregressive Integrated Moving Average methods to predict the total number of COVID-19 cases in the upcoming days in Bangladesh. We tuned both the models with different configuration parameters, and made 3 distinct configurations for each. After that, we applied all the different configurations on the same dataset, and the results were compared against each other in terms of statistical performance measures such as Mean Absolute Percentage Error (MAPE), Root Mean Squared Relative Error (RMSRE), Root Mean Squared Relative Error (RMSRE).