Using exponential random graph (p∗) models to generate social networks in artificial society

Abstract

Artificial society, which is a bottom-up method, has become a significant mean of studying complexity and complex phenomena in human society. Social networks play an important role in the research of social interaction among people, and are also key components of the artificial society. A good social network model should be both estimable and representable. Exponential random graph (p*) models (ERGMs) can satisfy the requirements. In this paper, ERGMs are applied to the generation of social networks in the artificial society, and a general process of generating social networks is proposed. As a case study, friendship networks in an artificial classroom are generated based on the statnet suite. The results indicate that ERGMs are efficient to generate social networks, and this method is practicable and worthy of application.