Complex network modelling and analysis of dengue epidemic in Selangor, Malaysia

Abstract

There are many examples of complex systems in the world from different domains of life. For instance, the World Wide Web, electric power grids, scientific collaboration networks, social network of friendships, and network of diseases such as HIV/AIDS. These systems can be better analyzed by converting them into complex networks. Mostly, complex networks are dynamic in natures which grow by adding new links and nodes. Nodes can represent elements of the systems and links depicts the interacting patterns between these elements. The dengue epidemic is a dynamic and complex phenomenon which has gained much attention due to its harmful effects that sometimes become a cause of death of a person. This problem is difficult to understand by just observing separate components which constitute this network. For this, modelling the way these units are interconnected with each other can enhance the understanding towards this epidemic as a whole. In this study, the dengue spreading phenomenon is addressed from the perspective of complex network and modelled using the dataset of weekly dengue affected cases in Selangor, Malaysia using scale-free network theory. Further, the dengue epidemic network is formalized and analyzed by projecting it from two-mode to one-mode network using three projection methods. Various network analysis metrics have been utilized in this study (such as: Closeness centrality, Betweenness centrality, Degree measures, Short path-length, and Eigenvector centrality). It is believed that this modelling and analysis will considerably help to comprehend the complex nature of the dengue epidemic network. This research reveals the focal nodes of the dengue virus, and the most crucial time is also highlighted when dengue virus is on its peak. Furthermore, we mathematically modelled the influence of the external and internal factors which become the cause of the dengue virus diffusion, and some exogenous factors have been shown that supressed the Aedes aegypti (dengue vector) network. Moreover, results showed that the dengue epidemic network comprises a scale-free network features which negates the theory of randomness. So, proper precautionary measures can be taken to prevent or reduce this disease. All the outcomes are valuable for the health officials and decision makers who deal with the epidemics.