Optimizing Hammerstein-Wiener Model for Forecasting Confirmed Cases of Covid-19

Noise poses challenge to nonlinear Hammerstein-Wiener (HW) subsystem model application, because HW subsystem need large number of parameter interactions. However, flexibility, soft computing, and automatic adjustment to dynamic observation for best model fitting make it potential for forecasting nonlinear data. In this article, we adopted improved HW inference from Levenberg-Marquardt optimization algorithm to optimize HW subsystem and to select best model parameters. Therefore, the adopted model is tested on COVID-19 confirmed reported cases, to estimate transmission rate of COVID-19 virus for period from 15th March 2020 to 29th April 2020. Model validation is carried out on small dataset, which outperforms some existing models. The adopted model is further evaluated using statistical metrics and reported best accuracy of 0.127 and 0.998 for Mean Absolute percentage error (MAPE) and coefficient of determination (R2) respectively, with best model complexity of 1.86. The obtained results are promising enough in predicting spread of COVID-19 virus and may inspire as guidance to relax lockdown restriction policies. © 2022, IAENG International Journal of Applied Mathematics. All Rights Reserved.

Data-Driven AI-Based Parameters Tuning Using Grid Partition Algorithm for Predicting Climatic Effect on Epidemic Diseases

Adaptive Neuro-fuzzy Inference System (ANFIS) remains one of the promising AI techniques to handle data over-fitting and as well, improves generalization. Presently, many ANFIS optimization techniques have been synergized and found effective at some points through trial and error procedures.In this work, we tune ANFIS using Grid partition algorithm to handle unseen data effectively with fast convergence. This model is initialized using a careful selection of effective parameters that discriminate climate conditions;minimum temperature, maximum temperature, average temperature, windspeed and relative humidity. These parameters are used as inputs for ANFIS, whereas confirmed casesof COVID-19 is chosen as dependent values for two consecutive months and first ten days of Decemberfor new COVID-19 confirmed cases according to the Department of disease control (DDC) Thailand. Theproposed ANFIS model provides outstanding achievement to predict confirmed cases of COVID-19 with $R{2} of 0.99. Furthermore, data set trend analysis is done to compare fluctuations of daily climatic parameters, to satisfy our proposition, and illustrates the serious effect of these parameters onCOVID-19 epidemic virus spread. © 2013 IEEE.