Predicting Newcomer Integration in Online Knowledge Communities by Automated Dialog Analysis

Abstract

Using online knowledge communities (OKCs) from the Internet as informal learning environments poses the question how likely these communities will be to integrate learners as new members. Such prediction is the purpose of the current study. Based on the approaches of voices interanimation and polyphony, a natural language processing tool was employed for dialog analysis in integrative versus non-integrative blog-based OKCs. Four dialog dimensions were identified: participants’ long-term persistence in the discourse, the community response to their participation, their communicative centrality, and their communicative peripherality. Hierarchical clusters built upon these dimensions reflect socio-cognitive structures including central, regular, and peripheral OKC members. While the socio-cognitive structures did not make a significant difference, integrative OKCs display significantly stronger peripherality, community response, and centrality as compared to non-integrative OKCs.