Randomness and clustering of responses in online learning networks
CIIT '07 The Sixth IASTED International Conference on Communications, Internet, and Information Technology
| Hi-index | 0.00 |
Online communities are described in terms of collections of virtual neighborhoods, each of which is a sub-set of interdependent members. The significant virtual neighborhoods are revealed by fitting parametric Markov Field models (p*) to the response relations of the communities. The underlying theoretical mechanisms are then deduced by matching the revealed virtual neighborhoods with the predictions of network emergence theories. We demonstrate that the underlying mechanisms are related to specific design features of the communities. This method can be extended to other relations in online communities and to longitudinal analysis, and applied to real-time monitoring of online communications.