To search, Click below search items.


All Published Papers Search Service


Data Analytic for Student Behaviour Classification in Social Learning Network


Andi Besse Firdausiah Mansur, Norazah Yusof, and Nada Omar Bajnaid


Vol. 18  No. 10  pp. 52-58


Social Learning network has grown tremendously over past year. This trend has stimulated E-learning system to evolve as social learning network. Meanwhile, the conservative clustering approach on social learning network still less explored due to its association relationship between students are exist. This paper proposed an approach to apply data analytic for revealing the behaviour of student from wiki activities using two major clustering techniques: hierarchical and Minimum Spanning Trees (MST) clustering. The first step is creating the matrix adjacency to determine indegree and outdegree level, and then convert the matrix into geodesic distance matrix. Meanwhile MST created by graph with weighted edge based on the intensity of relationship. The experimental study has shown that student who are possessed the high outdegree value tends to be at the same cluster. The other indication is high outdegree will influence the path in the network. Therefore, it will create shortest route in network, which is indicate the information passed through this student will be broadcasted instantaneously. As shown in the result section student1 and student12 is more active compared to others student. Our approach has been successfully classifying the active and passive student based on their participation on Moodle E-Learning Wiki. This means data analytic using hierarchical clustering and minimum spanning tree can be used to identify student performance or behaviour during study. Consequently, teacher may use it as reference for better e-learning system in the future.


Data Analytic, Social Learning Network, Hierarchical Clustering, Minimum spanning Trees, E-Learning.