Modelling user Behaviour in online Q&A communities for customer support

Abstract

With the increased popularity of Questions and Answers (Q&A) platforms, especially as a means to efficient customer support management, a lot of research has been carried out in order to study the user behaviour on Q&A sites. However, many research questions remain unanswered, as the underlying dynamics of replying in online communication platforms are not yet fully understood. One reason for this is that the interaction patterns in typical datasets with thousands of users and millions of posts are too complex to be broken down to the level of the individual users. In this paper, we present an agent-based model of online Q&A communities that is able to explain how these complex behaviour patterns evolve from the basic interactions of the individual agents. We evaluate our model on the SAP Community Network, and find that it closely reproduces Q&A behaviour of the real data.