An efficient neural network approach for the susceptible, infected, recovered and cross-immune delay differential system

The numerical computational performances of epidemic delay differential susceptible, infected, recovered, and cross-immune (SIRC) systems based on the coronavirus (CoV) are presented in this study. These computing frameworks are proposed using artificial neural networks (ANNs) enhanced with the Levenberg–Marquardt backpropagation (LBMBP), i.e., ANNs-LBMBP. Three different cases of the SIRC epidemic delay differential system have been numerically solved by using the designed computational structure. The perfection of the proposed ANNs-LBMBP approach is identified through the comparison of the accomplished and reference (Adam method) results. The results achieved by the epidemic SIRC delay differential model have been proposed by using training, corroboration, and test data performance to reduce the mean square error. For the efficiency and competence of the ANNs-LBMBP, a comprehensive investigation is accessible using the transition of state values, histogram error studies, regression data, and correlation.