Community building by evaluating cliques in graphs generated from the bipartite view of interaction graphs in GraphML format. Application in distance learning platforms

In the world of the Web, and more specifically Web 2.0, the precise knowledge of the profile and specific needs of users for monitoring or even for commercial purposes has become a necessity. To achieve this goal, a significant amount of information is generated and stored by Information Systems. Based on this massive source of information, the new challenge could be the creation of groups or networks of people sharing the same interests. Several approaches are possible. Our approach will be to form/build communities by performing clique determinations in the graphs generated from the bipartite view of the graph created by the interactions found in the platforms. The notion of cliques in the graphs makes it possible to determine the strongly linked elements [14], [15]. It is this concept applied to the users that we want to prove/demonstrate in our study. The aim/purpose of this work is to design a powerful tool for generating and adapting social groups. We extract tracking information of individuals and their collaboration/interaction history/experience from a platform in order to create or reorganize social groups.. There are two types of interaction information to be studied and characterized, namely, tracking information and collaboration information. To achieve our goal, we used a remote learning platform with standard elements of interactions in order to draw the expected conclusions.