Artificial neural network procedures for the waterborne spread and control of diseases

Abstract

In this study, a nonlinear mathematical SIR system is explored numerically based on the dynamics of the waterborne disease, e.g., cholera, that is used to incorporate the delay factor through the antiseptics for disease control. The nonlinear mathematical SIR system is divided into five dynamics, susceptible X(u), infective Y(u), recovered Z(u) along with the B(u) and Ch(u) be the contaminated water density. Three cases of the SIR system are observed using the artificial neural network (ANN) along with the computational Levenberg-Marquardt backpropagation (LMB) called ANNLMB. The statistical performances of the SIR model are provided by the selection of the data as 74% for authentication and 13% for both training and testing, together with 12 numbers of neurons. The exactness of the designed ANNLMB procedure is pragmatic through the comparison procedures of the proposed and reference results based on the Adam method. The substantiation, constancy, reliability, precision, and ability of the proposed ANNLMB technique are observed based on the state transitions measures, error histograms, regression, correlation performances, and mean square error values.

Keywords

• sir system
• delay term
• artificial neural network
• stochastic procedure
• levenberg-marquardt backpropagation