A radial basis scale conjugate gradient neural network process for the Zika model with human movement and reservoirs

The purpose of current research is to find the numerical solutions of the nonlinear Zika model with human movement and reservoirs (ZMHMR) by designing a novel radial basis scale conjugate gradient neural network (RB-SCGNN). This nonlinear model contains ten different groups, and the numerical solutions are presented by the stochastic RB-SCGNN process. A design of dataset is presented through the Runge-Kutta scheme to lessen the values of the mean square error by splitting the data into training as 72 %, while 14 %, 14 % for both verification and testing. Fifteen neurons in the hidden layers, single input, and radial basis activation function are used to solve the ZMHMR. The accuracy of the proposed scheme is judged through the overlapping of the outputs, whereas smaller values of the absolute error indicate the exactness of the RB-SCGNN. Additionally, the statistical representations using different operators validate the approach's trustworthiness.