Network Simulations

Social network simulation (SNS) is an emergent area of research that combines social network analysis and simulation, typically agent-based simulation. This area [Page 612]is often referred to as dynamic network analysis, as much of the focus of the combined modeling approach is on how networks evolve, change, and adapt. Additionally, SNS has a focus on how individual and group learning and behavior is impacted by and impacts the changes in the networks in which the individuals are embedded. Frequently, in social network simulations, the social network and other networks, such as the knowledge network, and/or the individuals or “nodes” are coevolving as agents interact, learn, and engage in various activities. The need to address complex systems but produce realistic results means that these SNSs typically focus on many types of networks simultaneously, not just the social networks. An example of such a model might be one that explores how communicating new ideas via diverse social media has differential impact on the movements of ideas and diseases through the population, and response to the information and disease by the populace.

There are various types of social network simulations; each has a unique perspective on the problem at hand, and each has its own collection of strengths and weaknesses. It is helpful to begin with more formal approaches that rely heavily on statistics and mathematical formalisms and then move on to less formal, bottom-up approaches. System dynamics is a top-down, aggregate view of networks. Regression or econometric approaches like quadratic assignment procedure provide a nonparametric approach to modeling dynamic social networks. More traditional parametric statistical approaches to SNS use methods such as expectation maximization or maximum-likelihood estimation to find the optimal (or near optimal) model parameters given the data. Finally, an agent-based SNS provides an intuitive, bottom-up approach for investigating social systems.

Regardless of the method used for social network simulation, there are unique sets of challenges around validation, analysis, prediction, and computational efficiency that are common to all. As each of these challenges are met and overcome, or sometimes sidestepped, SNS will only grow in its popularity and utility to both science and business worldwide.