Preferential response mechanism in online learning networks

In this research we ask whether actors in online learning choose their response partners at random. If not, we would like to discover what mechanism underlies the behavior of the network. To capture the complex feature space of the networks we map them into a high-dimensional feature space. A Multi-way Support Vector Machine algorithm is used to classify 35 observed response networks of online learners into a set of 5 representative stochastic network generation models. The result shows that all the response networks were classified to a preferential response model in which actors tend to respond to partners who are a-priori equipped with response attraction power. We provide a possible explanation for this behavior, based on the nature and goal of the online learning networks, and discuss ways in which the study can continue.