Comparative Analysis of Community Detection Methods for Online Learning Environments

The understanding, analysis, and prediction of the behaviours and dynamics of networks associated with several disciplines in sociology, criminology, biology, medicine, communication, economics, and academics have advanced significantly because of the discovery of community structure. Finding communities and grouping them together is a useful first step in figuring out the behavioural patterns and structural characteristics of social networks. Many educational approaches have recently gradually embraced online learning, which raises several concerns regarding how to evaluate students’ participation, teamwork, and behaviours in the brand-new, emergent learning communities. In this research work we have applied five algorithms of community detection on three different data set related to social network and online learning environment. The purpose of this work is to evaluate how community detection techniques are applied to network structure analysis in online learning environments. Experimental results indicate that the Girvan-newman and Louvain algorithm are giving the better results for community detection.