Integrated intelligent neuro-evolutionary computing approach to study SEIRC model representing campylobacteriosis transmission dynamics

The objective of the presented study is to develop a neuro-evaluation-based algorithm for the mathematical solution of the SEIRC model that describes the dynamics of campylobacteriosis transmission (CBT) using the artificial neural network along with log-sigmoid as an activation function (ANN-LS) and the properties of both genetic algorithms and sequential quadratic programming (GASQP). Five groups compose the mathematical SEIRC framework for the CBT model: susceptible (S), exposed (E), infected (I), recovered (R), and the bacterial population (C). The objective function is expressed in the mean squared error sense in the proposed ANN-LS-GASQP approach employing the approximate differential mapping of ANN for the SEIRC model. To evaluate the reliability, accuracy, and consistency of the proposed methodology, the designed intelligent solver ANN-LS-GASQP is compared with the state of the art Adam numerical solver. The solver has successfully achieved the absolute error in excellent ranges, which confirms the effectiveness and reliability of the solver ANN-LS-GASQP. The effectiveness of the suggested stochastic approach is examined using 100 trials with 5 neurons each, along with extra statistical indicators.