- Published on 18 September 2015
The power of any kind of network approach lies in its ability to simplify a complex system so as to better understand its function as a whole. Sometimes it is beneficial, however, to include more information than is available in a simple graph of nodes and links. Adding information related to the timing of interactions can facilitate more accurate predictions, as well as a deeper mechanistic understanding.
The drawback, however, is that there are not many methods available for the study of temporal networks, partly because it is a relatively young field, and partly because it is inherently more difficult to develop such methods as compared to those for static networks. In this EPJ B colloquium, Petter Holme reviews the methods that are currently available to analyse and model temporal networks and the processes taking place within them, with a focus on methods developed within the last three years. This includes models of the spread of infectious disease, opinions and rumors in social networks, information packets in computer networks, as well as various types of signaling in biology, and more.
Finally, the paper features a discussion of the future directions for temporal network research in which the author anticipates that new applied areas will discover temporal networks as a modelling framework. Certainly new ideas will be welcome for the visualization of temporal networks and more work will be needed on how to resample original data, or scale up results and extrapolate from e.g. spreading studies on empirical networks to larger populations.
Petter Holme (2015),
Modern temporal network theory: a colloquium,
European Physical Journal B, DOI: 10.1140/epjb/e2015-60657-4