Scale-free networks

The widespread presence of networked systems in technical applications as well as in the business world makes network research useful for future applications in information systems (IS) research. Research into networks means not just analyzing the topology of networks, but also examining the dynamics of the processes that take place in them. Research into diffusion processes in networks has been shown to provide especially fruitful insights into several kinds of IS domains such as e.g. information technology (IT) robustness in system failure situations, denial-of-services attacks, and computer virus spreading. For decades network research was dominated by random networks, also called Erdös and Renyi networks or just ER-networks [ErRe59]. ER-networks are random in the sense that the links between the nodes of a network are created with the help of a random variable. Recently, following the introduction of what were called “scale-free” networks by Barabasi and Albert [BaAl99], the field has received growing attention from scientific research. Scale-free networks seem to match real world applications much better than ERnetwork models [Bara03]. The term scalefree refers to the distribution principle of how many links there are per node. This article is intended to provide an introduction into the metrics of modern network analysis. The basic network measures like the degree distribution, the clustering coefficient and the average path length will be presented to enable a better understanding of the two network models “ER” and “scale-free”. Diffusion processes in networks depend very much on network topology. Both network models will be examined in terms of their different diffusion behavior. Finally, the potential impact for real-life applications will be shown by the introduction of a new application of scalefree networks: The simulation of the implications of network topology for the price building process in a security trader network that relies on a communication network.