Global dynamics and numerical simulation of a modified epidemiological model for viral marketing on social networks

Abstract

The aim of this work is to conduct a rigorous mathematical analysis for global dynamics and numerical simulation of a recognized viral marketing (VM) model, which is described by a system of ordinary differential equations (ODEs). We first establish positivity and boundedness of solutions and then investigate local and global asymptotic stability properties of possible equilibrium points. As an important consequence, complex dynamics of the VM model is determined fully.

Secondly, we develop the Mickens’ methodology to design a nonstandard finite difference (NSFD) scheme, which is useful in numerical simulation of the VM model. The main advantage of the constructed NSFD scheme is that it has the ability to preserve important mathematical features of the continuous-time model for all finite values of the step size. These features include the positivity and boundedness of solutions, positively invariant sets, equilibrium points and their asymptotic stability properties. Consequently, the NSFD scheme is not only effective to simulate dynamics of the VM model, but also easy to be implemented.

Thirdly, to emphasize implications of the constructed mathematical analysis, an extended version combining the integer-order ODE model under consideration with the Caputo fractional derivative is considered and analyzed. From the mathematical analysis performed for the integer-order VM model, global dynamics of the fractional-order VM model is also investigated rigorously.

Finally, the theoretical insights are supported by a set of illustrative numerical experiments.

The findings of this research not only improve some existing results in the literature, but may also provide several useful real-life applications.