A stochastic neural network for the numerical solutions of the nonlinear fractional order Zika virus model using reservoirs and human motion

Abstract

The current research investigations present the solutions of the fractional order Zika virus model using reservoirs and human motion, which has ten different classes. The fractional order Caputo derivative is used to get more precise solutions of the mathematical nonlinear model. An artificial neural network process is presented using the reliable Bayesian regularization approach, sigmoid activation function, twenty neurons along with different testing and training data. The design of the dataset is achieved through the Adam technique, which is used to lessen the mean square error by distributing the statistics for training (78 %), authentication (10 %), and testing (12 %). Three different cases of the model using the fractional order values are accessed to check the correctness of the designed scheme based on the overlapping of results and negligible absolute error. Furthermore, the statistical representations have been implemented to substantiate the consistency of the solver.