This research paper is to investigate computationally the stochastic SIS epidemic model including the coexistence of two diseases, based on the system of stochastic differential equations. A well-known Levenberg–Marquardt method of backpropagation with neural networks is used to approximate the considered epidemic model with the help of algorithms designed using an artificial intelligence approach. This epidemic model is comprised of three main classes: S(t) stands for susceptible population under consideration, X(t) and Y (t) stand for infected population due to virus A and virus B, respectively, at time t. To obtain the specific datasets which are to be used in the designed technique, the Wiener process is applied for a large number of scenarios by varying; the inhibitory effect which is caused by an increase in infected individuals, death rate due to virus, the rate at which infected individuals come in contact with susceptible individuals, rate of recovery after treatment, the white noise and the natural mortality rate. For obtaining the approximate solution of the considered model which is in good agreement with the referenced dataset, we have used our technique with a selection of arbitrary data from the referenced dataset as training, validation, and testing samples. The computational results, which are obtained in the form of performance plots, mean square error, fitness plots, regression index and error histogram provide evidence that the proposed technique is consistent, competent and accurate.
Intelligent computing paradigm for stochastic SIS model with double epidemic diseases
Haider A.;
2025-01-01
Abstract
This research paper is to investigate computationally the stochastic SIS epidemic model including the coexistence of two diseases, based on the system of stochastic differential equations. A well-known Levenberg–Marquardt method of backpropagation with neural networks is used to approximate the considered epidemic model with the help of algorithms designed using an artificial intelligence approach. This epidemic model is comprised of three main classes: S(t) stands for susceptible population under consideration, X(t) and Y (t) stand for infected population due to virus A and virus B, respectively, at time t. To obtain the specific datasets which are to be used in the designed technique, the Wiener process is applied for a large number of scenarios by varying; the inhibitory effect which is caused by an increase in infected individuals, death rate due to virus, the rate at which infected individuals come in contact with susceptible individuals, rate of recovery after treatment, the white noise and the natural mortality rate. For obtaining the approximate solution of the considered model which is in good agreement with the referenced dataset, we have used our technique with a selection of arbitrary data from the referenced dataset as training, validation, and testing samples. The computational results, which are obtained in the form of performance plots, mean square error, fitness plots, regression index and error histogram provide evidence that the proposed technique is consistent, competent and accurate.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
