A graph clustering model to predict student retention using DAG

Educational data mining is one of the prominent areas to explore information in the field of education. One of the major issues faced by universities nowadays is retention. Out of the many factors that influence this, curriculum structure has not received much attention. This paper models each student data as a Directed Acyclic Graph and applies the sub-graph mining method followed by the graph clustering method. The experimental results show that the proposed model gives an accurate and efficient way for predicting a student’s chance of retention in a curriculum path.