The numerical methods such as the artificial neural networks with greater probability and nonlinear configurations are more suitable for estimation and modeling of the problem parameters. The numerical methods are easy to use in applications as these methods do not require costly and time-consuming tests like the experimental study. In this study, we use the Levenberg–Marquardt-based backpropagation Process (LMP) to create a computing paradigm that makes use of the strength of artificial neural networks (ANN), known as (ANN-LMP). Here we use the ANN-LMP to obtain the solution of the second-grade fluid in a rotating frame in a porous material with the impact of a transverse magnetic field. The 1000 data set points in the interval [0,1] are used for the network training to determine the effect of various physical parameters of the flow problem under consideration. The experiment is executed of six scenarios with different physical paramaters. ANN-LMP is used for evaluating the mean square errors (MSE), training (TR), validation (VL), testing (TT), performance (PF) and fitting (FT) of the data. The problem has been verified by error histograms (EH) and regression (RG) measurements, which show high consistency with observed solutions with accuracy ranging from E-5 to E-8. Characteristics of various concerned parameters on the velocity and temperature profiles are studied.
