The structural evolution of an online discussion network

A discussion network is an important kind of social network, and it has been researched by many scholars in recent years. In this paper, we mainly studied its structural evolution based on an empirical study of a famous online discussion that happened in China in 2008. We found that the scale growth of the network shows an S shape, the degree distribution represents the power law in the first halfway, and the network shows a degree of disassortativity characteristic. We also classified the most active participants into different groups by their opinions and studied the structural evolution of opposite groups. It was observed that the evolution of nodes in each group were very similar, but the evolution of densities were obviously different. Specifically, we found that the most active participants preferred to converse with those belonging to the opposite groups at the beginning and tended to converse with anybody regardless of group as the discussion network grew. In the paper, these evolution patterns are revealed, and future lines of research are also considered.